Topic_Modeling_AirbnbFlorence.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "71d92e0f-a5ff-4dbe-a8bd-05b30e7dd471",
   "metadata": {},
   "source": [
    "# Q2. What are the top topics in guest reviews?\n",
    "\n",
    "## Table of Contents: \n",
    "\n",
    "\n",
    "* [Topic Modeling: Latent Dirichlet Allocation (LDA)](#chapter4_)\n",
    "    * [Cleaning and Pre-processing data for LDA](#chapter4_1)\n",
    "    * [Train LDA Topic Model](#chapter4_2)\n",
    "        * [Select the Optimal Number of Topics](#chapter4_2_2)\n",
    "        * [Build LDA Model and Interpret Results](#chapter4_2_3)\n",
    "        * [Evaluate LDA Model](#chapter4_2_4)\n",
    "    * [Visualize topic-keyword of LDA](#chapter4_3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8028511d-9033-4dec-93ae-3119b5971bc1",
   "metadata": {},
   "source": [
    "### Import Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "1f709a3c-3eb7-4054-a9d7-36c6bfdd3b9a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "import seaborn as sns\n",
    "sns.set_theme(style=\"whitegrid\")\n",
    "\n",
    "\n",
    "import spacy\n",
    "from spacy.lang.en.stop_words import STOP_WORDS\n",
    "from spacy.lang.en.examples import sentences\n",
    "\n",
    "import pickle \n",
    "\n",
    "from nltk.tokenize import word_tokenize\n",
    "\n",
    "\n",
    "import gensim\n",
    "from gensim.utils import simple_preprocess\n",
    "import gensim.corpora as corpora\n",
    "from gensim.models import Phrases,CoherenceModel, LdaModel\n",
    "\n",
    "import pyLDAvis.gensim_models\n",
    "import pyLDAvis\n",
    "import pyLDAvis.sklearn\n",
    "\n",
    "#from imp import reload\n",
    "import importlib\n",
    "\n",
    "import time\n",
    "\n",
    "from tabulate import tabulate\n",
    "import re"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1e9dd602-e29f-4175-a3ba-a8ce9d2d7550",
   "metadata": {},
   "source": [
    "[Back to TOC](#Table-of-Contents:)\n",
    "<a class=\"anchor\" id='chapter4_'></a>\n",
    "# Topic Modeling: Latent Dirichlet Allocation (LDA)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "74ce8447-a5c7-4bf3-8e52-7163589974f5",
   "metadata": {},
   "source": [
    "<a class=\"anchor\" id='chapter4.1'></a>\n",
    "## Cleaning and Pre-processing data for LDA"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cc9602ba-9940-4542-aaf3-3c6c13dd1465",
   "metadata": {},
   "source": [
    "For this analysis, we further clean and pre-process reviews following these steps:<br>\n",
    "1. Lemmatize <br>\n",
    "2. Remove punctuation <br>\n",
    "3. Remove stop words <br>\n",
    "4. Remove numbers <br>\n",
    "5. Keep words with length > 2 <br>\n",
    "6. Tokenize <br>\n",
    "\n",
    "Then, we convert text into a bag of words as a dictionary, where to each word (key) is associated a numeric id (value). And finally, we count the number of times each word occurs within the corpus.\n",
    "\n",
    "Packages required: <br>\n",
    "- spacy\n",
    "- nltk.tokenize\n",
    "- gensim.corpora"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "90ded0f4-ec9f-47c2-8c16-1bf190d2dbf8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Upload data\n",
    "\n",
    "df_reviews_sample_sentiment_bitrigrams = pd.read_pickle('./data/df_reviews_sample_sentiment_bitrigrams.pkl')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e72de1b7-10ea-4d93-84e0-7006722cdd58",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>listing_id</th>\n",
       "      <th>id</th>\n",
       "      <th>date</th>\n",
       "      <th>reviewer_id</th>\n",
       "      <th>reviewer_name</th>\n",
       "      <th>comments</th>\n",
       "      <th>length_comments</th>\n",
       "      <th>language</th>\n",
       "      <th>comments_english</th>\n",
       "      <th>polarity_compound</th>\n",
       "      <th>sentiment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>39978</th>\n",
       "      <td>735370</td>\n",
       "      <td>151499476</td>\n",
       "      <td>2017-05-13</td>\n",
       "      <td>107924476</td>\n",
       "      <td>Jutta</td>\n",
       "      <td>eine schön gelegene wohnung in der etage die w...</td>\n",
       "      <td>59</td>\n",
       "      <td>de</td>\n",
       "      <td>a beautifully located apartment on the same fl...</td>\n",
       "      <td>0.0258</td>\n",
       "      <td>neu</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>196555</th>\n",
       "      <td>5909451</td>\n",
       "      <td>142261214</td>\n",
       "      <td>2017-04-07</td>\n",
       "      <td>118986470</td>\n",
       "      <td>Michael</td>\n",
       "      <td>alles bestens sehr gute lage nahe der piazza d...</td>\n",
       "      <td>25</td>\n",
       "      <td>de</td>\n",
       "      <td>everything was very good, very good location n...</td>\n",
       "      <td>0.8803</td>\n",
       "      <td>pos</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>340512</th>\n",
       "      <td>14976201</td>\n",
       "      <td>437134397</td>\n",
       "      <td>2019-04-14</td>\n",
       "      <td>129853528</td>\n",
       "      <td>Kristine</td>\n",
       "      <td>das airbnb hat uns sehr gut gefallen wir haben...</td>\n",
       "      <td>120</td>\n",
       "      <td>de</td>\n",
       "      <td>We liked the airbnb very much, although we onl...</td>\n",
       "      <td>0.9154</td>\n",
       "      <td>pos</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>189256</th>\n",
       "      <td>5745043</td>\n",
       "      <td>658966293</td>\n",
       "      <td>2020-08-31</td>\n",
       "      <td>160489786</td>\n",
       "      <td>Mark</td>\n",
       "      <td>perfekt für einen urlaub in florenz und der an...</td>\n",
       "      <td>15</td>\n",
       "      <td>de</td>\n",
       "      <td>perfect for a vacation in florence and traveli...</td>\n",
       "      <td>0.5719</td>\n",
       "      <td>pos</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>149460</th>\n",
       "      <td>4143454</td>\n",
       "      <td>179633637</td>\n",
       "      <td>2017-08-08</td>\n",
       "      <td>97926976</td>\n",
       "      <td>Uwe</td>\n",
       "      <td>wir hatten als familie  pesonen eine sehr schö...</td>\n",
       "      <td>96</td>\n",
       "      <td>de</td>\n",
       "      <td>we had a very nice time as a family pesonen in...</td>\n",
       "      <td>0.9432</td>\n",
       "      <td>pos</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        listing_id         id        date  reviewer_id reviewer_name  \\\n",
       "39978       735370  151499476  2017-05-13    107924476         Jutta   \n",
       "196555     5909451  142261214  2017-04-07    118986470       Michael   \n",
       "340512    14976201  437134397  2019-04-14    129853528      Kristine   \n",
       "189256     5745043  658966293  2020-08-31    160489786          Mark   \n",
       "149460     4143454  179633637  2017-08-08     97926976           Uwe   \n",
       "\n",
       "                                                 comments  length_comments  \\\n",
       "39978   eine schön gelegene wohnung in der etage die w...               59   \n",
       "196555  alles bestens sehr gute lage nahe der piazza d...               25   \n",
       "340512  das airbnb hat uns sehr gut gefallen wir haben...              120   \n",
       "189256  perfekt für einen urlaub in florenz und der an...               15   \n",
       "149460  wir hatten als familie  pesonen eine sehr schö...               96   \n",
       "\n",
       "       language                                   comments_english  \\\n",
       "39978        de  a beautifully located apartment on the same fl...   \n",
       "196555       de  everything was very good, very good location n...   \n",
       "340512       de  We liked the airbnb very much, although we onl...   \n",
       "189256       de  perfect for a vacation in florence and traveli...   \n",
       "149460       de  we had a very nice time as a family pesonen in...   \n",
       "\n",
       "        polarity_compound sentiment  \n",
       "39978              0.0258       neu  \n",
       "196555             0.8803       pos  \n",
       "340512             0.9154       pos  \n",
       "189256             0.5719       pos  \n",
       "149460             0.9432       pos  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_reviews_sample_sentiment_bitrigrams.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "1c9f9318-dce6-40f2-973e-b65c521af497",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(51047, 11)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_reviews_sample_sentiment_bitrigrams.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "78b1f529-01d8-4c24-89a0-d1ba24a9fa87",
   "metadata": {},
   "outputs": [],
   "source": [
    "nlp = spacy.load('en_core_web_sm', disable=[\"parser\", \"ner\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "7b0dc918-3b7f-4526-95c9-a0731a3569e1",
   "metadata": {},
   "outputs": [],
   "source": [
    "def pre_process_text(text):\n",
    "    \n",
    "    '''Clean and pre-process data.\n",
    "    '''\n",
    "    \n",
    "    cleaned_text = []\n",
    "    document = nlp.pipe(text, batch_size=256, n_process=6)\n",
    "    for doc in document:\n",
    "        text = [\n",
    "            token.lemma_\n",
    "            for token in doc\n",
    "            if not token.is_punct\n",
    "            and not token.is_stop\n",
    "            and not token.like_num\n",
    "            and token.is_alpha\n",
    "            and len(token.orth_) > 2\n",
    "        ]\n",
    "        cleaned_text.append(\" \".join(text).lower())\n",
    "    return cleaned_text"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "bddcb1bc-d775-47a9-a0c8-171382fc76fa",
   "metadata": {},
   "outputs": [],
   "source": [
    "def sentence_to_words(cleaned_text):\n",
    "    \n",
    "    '''Tokenize data\n",
    "    '''\n",
    "    \n",
    "    data_words = []\n",
    "    for sentence in cleaned_text:\n",
    "        item = word_tokenize(sentence)\n",
    "        data_words.append(item)\n",
    "    return data_words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "928e760f-ee8c-4f3f-a344-7e960c9cfdd8",
   "metadata": {},
   "outputs": [],
   "source": [
    "def prepare_data_words(data):\n",
    "    \n",
    "    '''Clean and pre-process data.\n",
    "    \n",
    "    Prepare Dictionary and Corpus for LDA.\n",
    "    \n",
    "    Dictionary: Unique ids for each unique word\n",
    "    Corpus: Word document frequency\n",
    "    '''\n",
    "    \n",
    "    # Pre-process text data\n",
    "    data_clean = pre_process_text(data)\n",
    "    \n",
    "    # Tokenize text data\n",
    "    data_words = sentence_to_words(data_clean)\n",
    "    \n",
    "    # Create Dictionary\n",
    "    id2word = corpora.Dictionary(data_words)\n",
    "    \n",
    "    # Filter extreme values\n",
    "    # no_below (int) filters out words appearing less than n times\n",
    "    # no_above (float) filters out words appearing in more than x% of all reviews\n",
    "    id2word.filter_extremes(no_below=5, no_above=0.1) \n",
    "  \n",
    "    # Create Corpus\n",
    "    texts = data_words\n",
    "    \n",
    "    # Term Document Frequency\n",
    "    corpus = [id2word.doc2bow(text) for text in texts]\n",
    "    \n",
    "    return id2word, corpus, data_words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "4f4087da-206b-4805-9981-12ff2a18645b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert reviews to list\n",
    "\n",
    "reviews_list = df_reviews_sample_sentiment_bitrigrams['comments_english'].to_list()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "85ce7395-7b32-47a9-ab4d-9c6cdb293d56",
   "metadata": {},
   "outputs": [],
   "source": [
    "id2word, corpus, data_words = prepare_data_words(reviews_list);"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7e8a91f2-1781-4b25-b8ef-3b5441cec728",
   "metadata": {},
   "source": [
    "[Back to TOC](#Table-of-Contents:)\n",
    "<a class=\"anchor\" id='chapter4_2'></a>\n",
    "## Train LDA Topic Model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d7b1395d-c98e-48c3-b5a6-1d834cf611c3",
   "metadata": {},
   "source": [
    "**Packages** required: <br>\n",
    "- gensim https://pypi.org/project/gensim/\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "162125e7-d8e8-4741-b6eb-0100e3194385",
   "metadata": {},
   "source": [
    "<a class=\"anchor\" id='chapter4.2.2'></a>\n",
    "### Select the Optimal Number of Topics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "13f80c2d-6cb6-4d6e-a8ae-0a23aeac8fba",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Round: 1\n",
      "--- 5.7380530834198 seconds ---\n",
      "Round: 2\n",
      "--- 19.069741010665894 seconds ---\n",
      "Round: 3\n",
      "--- 34.50109601020813 seconds ---\n",
      "Round: 4\n",
      "--- 51.365859031677246 seconds ---\n",
      "Round: 5\n",
      "--- 68.14608097076416 seconds ---\n",
      "Round: 6\n",
      "--- 84.78217005729675 seconds ---\n",
      "Round: 7\n",
      "--- 101.90255808830261 seconds ---\n",
      "Round: 8\n",
      "--- 119.27384901046753 seconds ---\n",
      "Round: 9\n",
      "--- 136.63148999214172 seconds ---\n",
      "Round: 10\n",
      "--- 154.5361888408661 seconds ---\n"
     ]
    }
   ],
   "source": [
    "# Optimize the number of topics\n",
    "start_time = time.time()\n",
    "coherence = []\n",
    "for k in range(1,11):\n",
    "    print('Round: '+str(k))\n",
    "    lda_model = gensim.models.LdaMulticore(corpus=corpus,\n",
    "                                       id2word=id2word,\n",
    "                                       num_topics=k,\n",
    "                                       chunksize=100,\n",
    "                                       passes=10,\n",
    "                                       iterations=200,\n",
    "                                       random_state=42,\n",
    "                                       per_word_topics=True)\n",
    "\n",
    "\n",
    "    \n",
    "    cm = gensim.models.coherencemodel.CoherenceModel(model=lda_model, texts=data_words,\n",
    "                                                     dictionary=id2word, coherence='c_v')\n",
    "    coherence.append((k,cm.get_coherence()))\n",
    "    ## Print time taken to train the model\n",
    "    print(\"--- %s seconds ---\" % (time.time() - start_time))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "f4a00433-4351-4c8f-b0df-04ab509a178f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAm0AAAFVCAYAAAC9ytjjAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjQuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8rg+JYAAAACXBIWXMAAAsTAAALEwEAmpwYAABj+ElEQVR4nO3deVhU5d8G8HvY931TUQRkEUFkUdEMcws3UlHT3NPMJaV8yyVzxRaXFA3FLEtzTcvtZ5lamqZpCpgsMqKAGqgMyKKArDPn/YOcHEEdZBkG7s91dcVZ5pzvzAG5Oc95nkckCIIAIiIiImrQNFRdABERERE9H0MbERERkRpgaCMiIiJSAwxtRERERGqAoY2IiIhIDTC0EREREakBhjaiKvz222+YNGkSunTpAh8fHwwePBg7d+5EWVlZtY4TEREBHx+fOqqyYZPJZNi9ezdCQkLQoUMH+Pj4YPjw4di7dy840lCF9PR0uLm5YcKECVVu/+STT9CzZ886r2PevHkYOHBgnZ+nOsRiMYKDg+Hp6YmpU6dW2u7m5vbc/9LT02tcR1P+GaaGR0vVBRA1NEuXLsX333+PwYMH44033oCBgQEuXryIFStW4K+//sLatWuhqamp6jIbvDVr1mDHjh14++234e3tjfLycpw/fx5LlizBrVu3MHv2bFWX2GCcP38e+/fvR0hIiKpLaTAiIyORm5uLL7/8Era2tpW279mzR/71zZs3MXfuXCxatAjt2rWTr7exsalxHcOHD0f37t1rfByi2sDQRvSYgwcPYteuXQgLC8OIESPk67t27QpXV1fMmjULhw8fxuDBg1VXpBooLS3Ftm3bMGPGDLz99tvy9d27d4dIJMJ3332HKVOmwMTERIVVNhzGxsZYsWIFunfvDktLS1WX0yDk5eXBw8MD3bp1q3J7hw4d5F/r6uoCANq0aaOwvjbY2dnBzs6uVo9J9KLYPEr0mG+++QZubm4Kge2R/v37Y+LEiTA3N5evS09Px7vvvouuXbvCx8cH06ZNw82bNyu99siRIwgKCoKXlxdCQkJw6dIlhe0JCQkYP348vL29ERAQgGXLlqGoqEi+fezYsVi4cCEmTZoEX19frFixAgCQnZ2NOXPmoFOnTvDx8cHUqVORlpYmf11ERARCQkLw008/yc8/dOjQSue/cOECRo8eDR8fHwQGBmL58uUoKSlRur4nFRQUoKSkpMpm0BEjRuDdd9+FTCZT+vxRUVEYPXo0fH190bVrV4SFhaGwsLDGn09V3Nzc8P3332PatGnw9vZGz549sWPHDoV9ysvLsW7dOrzyyivya3r+/HmF9/PoON26dUP37t2f2VQ3ffp0lJaW4pNPPnlmbWPHjsWUKVMU1m3duhVubm7y5Z49e+Krr77CggUL4Ofnh86dO+OLL75Afn4+PvjgA/j4+KBHjx7Yv39/peNv3rwZXbt2ha+vL95//31kZ2crbP/zzz8xfPhwtG/fHoGBgVi3bh2kUqnCuT///HO8/vrr8Pf3x9atW6t8H2VlZfjqq6/k35PBwcE4fPiwfLubmxsuXryI06dPw83NDRcuXHjm5/I0OTk5WLBgAQIDA+Ht7Y1x48YhPj5evn3//v3w8fHB2bNn0bdvX3To0AFjxoyBWCyW7/Nk86hUKsWXX36J3r17w9vbG4MGDcJvv/0m3x4bGyv/Xu7UqRNCQ0Nx+/btF6qf6EkMbUT/yszMxLVr157ZFDJ37lz59oyMDAwfPhy3bt3C4sWL8dlnnyE9PR2jRo2CRCKRv6aoqAjh4eEIDQ3FunXrUFRUhJkzZ6K8vBwAkJycjDFjxkAkEmHt2rX44IMPcOTIEbz33nsK596/fz/s7e3xxRdfoF+/figuLsa4ceMQExODBQsWYOXKlbh37x7GjBmD+/fvy1938+ZNfPHFF5gxYwYiIiJQUlKCd999V37+uLg4TJw4EcbGxggPD8fMmTPxww8/yAOEsvU9zsLCAp6enoiIiMDixYtx5swZechq3bo1Jk+eDDMzM6XOf/r0aYwbNw7W1tby7T///DOmTJmiEPxe9POpyueffw4DAwNERESgT58+WLZsGfbu3SvfvnDhQmzZsgXjxo3Dhg0b4OTkhMmTJ1cKw5GRkQgLC8OsWbNgb2//1PM1b94c7777Ln7++WecOnXqmbUp48svv4RUKsX69evRr18/bNiwAcOGDYO1tTXWrVsHJycnLFq0CHfu3JG/JjU1FXv37sWiRYuwaNEi/PXXX5g5c6Z8+/nz5zF58mTY29tj/fr1mDRpErZs2YKPP/5Y4dxbtmxBYGAgVq1ahcDAwCrrmzt3LiIjI/H6669j48aN8PHxwQcffIAffvgBQEXTp4eHB3x9fbFnzx6FJk9lFRYW4o033sC5c+fw/vvvIzw8HIIgYMyYMUhKSpLvV1paig8++ACjRo3CmjVr5N83TwbWRz777DOsX78eISEh+PLLL+Ht7Y3Q0FBER0ejqKgIb7/9NmxtbREZGYlly5YhMTER//d//1ft+omqJBCRIAiCEBsbK7i6ugq7du1Sav/PPvtM6NChg5CdnS1fl52dLfj4+AifffaZIAiC8MUXXwiurq7C33//Ld/n2LFjgqurqyAWiwVBEIRZs2YJPXv2FEpKSuT7REVFCa6ursLFixcFQRCEMWPGCB07dhRKS0vl++zevVto27atkJycLF+Xn58v+Pv7CxEREQrnj42Nle/z22+/Ca6urkJ8fLwgCIIwffp04dVXXxXKy8vl+2zfvl0ICQkRysvLlaqvKunp6cLIkSMFV1dXwdXVVWjbtq0wcuRIYffu3Qrnet75hwwZIrz++usKx/7jjz8EV1dX4cSJEzX6fKri6uoqDB8+XGHdzJkzhR49egiCIAjJycmCq6ursHfvXoV9xo0bJ4wdO1YQBEH466+/BFdXV2Hz5s1PPY8gCEJaWprg6uoq/PLLL0J5ebkQEhIidO/eXSgoKBAEQRA+/vhj+Xkfvc+3335b4RhbtmwRXF1d5cs9evQQ+vbtK8hkMkEQBKGoqEhwd3eX1yYIgnDr1i3B1dVV+PXXXwVBEIS5c+cKbdu2FVJTU+X7nDp1SnB1dRWioqIEQRCE119/XRg5cqTCuQ8cOCC4u7sLaWlp8nMPGDDgme/56tWrgqurq7B7926F9bNmzRICAgLk3wdVvdenSUxMFFxdXYW//vpLvm7btm2Cu7u7cP36dfm6kpISoUePHsKMGTMEQRCEffv2VfqZz87OFry9vRV+hjp06CAIgiDk5uYKbdu2FdavX69w/jFjxggbN26U/xty6dIl+bYLFy4I69atE6RSqVLvhehZeKeN6F+POhc8fvfmWaKiotC5c2dYWFjI11lYWKBLly64ePGiwnHbt28vX27RogUAID8/H0BFU9pLL70EDQ0NlJeXo7y8HB06dICRkZFCk1urVq2gra0tX75w4QIcHBzg4OAgf52enh78/Pzw119/yffT0tKCp6enfPnR8zmPmjf//vtvBAYGKnSuGDNmDPbt2wdNTU2l63tSixYtsHv3bhw6dAjvvfce/Pz8kJCQgMWLF2PChAkoLS197vmLi4uRmJiIvn37Khz75ZdfhqmpKaKiomr8+VSlf//+Csu9evXC7du3kZGRIb+2gYGB8uOWl5eje/fuuHTpkvx9ARXPWClLU1MTn3zyCbKyshAeHq7066rSvn17iEQiAICenh4MDQ0Vvgce3eV88OCBfJ2LiwscHR3ly4GBgdDW1kZcXByKiooQFxeHHj16KLznwMBAyGQyheZLZ2fnZ9YWHR0NAJWuaf/+/ZGTk4OUlJQXe9NPiIqKQps2bRSugY6ODnr37q3w8wkAAwYMkH9tYWGBDh06yOt8XGxsLKRSaaUevdu3b8fUqVPh5OQEMzMzTJ06FWFhYTh9+jQ6dOiA0NBQaGjw1y3VHDsiEP2rWbNmAIC7d+8+dZ/MzExYWVlBQ0MDDx48QNu2bSvtY2lpieTkZPmyrq6uwj/Yj75+FA7z8vKwZ88ehd5wj2RlZSkc93F5eXlITU2tsumodevW8q91dHSeef779+8/8+F3Zet7Gnd3d7i7u2PatGkoKCjA2rVrsX37dhw+fBhDhw595vnz8/MhCEKV2y0sLFBQUCBfftHPpypP9jp8FMzz8vKQl5cHAE9t+svNza30OmW5u7tjwoQJ+PbbbxEcHFyt1z7O0NCw0jp9ff1nvsbKykphWSQSwczMDJmZmXjw4AFkMhlWr16N1atXV3rts75Pn3T//n1oaWnJg+OT53/8mtbEgwcPKr2nR+d5/HlIXV3dSh1iLCwscOPGjUqvfdSs/rTramRkhB07dmDDhg04cOAAdu7cCRMTE8yaNQujRo2qydshAsDQRiRnYWEBDw8PnDlzBh988EGV+7z55puwsrLCd999B1NTU9y7d6/SPvfu3av0C+lZjIyM0KtXL7zxxhuVtj3e6eFJxsbGcHd3r/RMEVAR1Kpz/pycHIV1eXl5uHLlCnx9fV+ovq1bt+Kbb77BqVOnFO6gGRkZ4aOPPsLhw4fld1SedX5vb2+IRKIqny963udck8/n8eAFQH5+CwsLGBsbQyQSYffu3dDSqvxPqLm5eZWdUZQ1c+ZMHD9+HAsWLIC/v3+l7U/eCX748OELn+txTz7nJ5PJkJubCzMzM3kInDZtGnr16lXptdUZWsPU1BTl5eXIy8tTuH6Pfpaq87PzvPOkpqZWWp+VlaVwjpKSEhQVFSmE2uzs7CqDmbGxMYCK74/HhyERi8UQBAEeHh5wcXHB2rVrUVpaipiYGHz33XdYunQp2rVrB29v71p5b9R08X4t0WPGjx+Pq1evyh+IftyhQ4eQnJwsvwPi5+eHCxcuKASOnJwcnD9/Hr6+vkqf08/PD6mpqfD09ISXlxe8vLzQrFkzrF69GtevX3/q63x9fZGeno4WLVrIX+fp6YmtW7dW62F2Hx8f/PHHHwph4MiRI5gyZQqkUukL1efo6IjMzEz8+OOPlbZlZmaisLAQrq6uzz0/ALRt2xZHjx5VOMaZM2eQn5//zM+5Jp/Pk9tPnDgBJycn2NjYwM/PD4IgoLCwUH5cLy8vnD9/Hlu3bq0yyFWHnp4eli5dimvXruF///ufwjYjIyNkZmYqrIuJianR+R65evWqwh8hJ06cQHl5OTp16gQjIyO4u7sjLS1N4T1ra2tjzZo1yMjIUPo8fn5+AFDpmh45cgSWlpbPvQtanfMkJycrNLeWlpbit99+q/R98/vvv8u/zs7OxuXLl9G5c+dKx2zfvj20tLQU9geARYsW4ZtvvsEff/yBLl26ICcnBzo6OujSpQsWLlwIAAqdPoheFO+0ET1m0KBBOHXqFBYtWoS4uDj06tULIpEIZ8+exe7du9GvXz8MHToUADBhwgQcOHAAEydOxPTp0yEIAjZu3AgdHR2MHz9e6XNOnz4dI0eOxLvvvouhQ4eitLQUkZGRuHv3Ljw8PJ76umHDhmH79u2YOHEi3n77bZiZmWHPnj04fvw4XnvtNaXPP3XqVIwePRqhoaF4/fXXkZGRgbVr12LMmDEwMjJ6ofoCAwPRu3dvLF26FFeuXMErr7wCY2NjJCcn49tvv0Xbtm3lz4097/wzZ87E9OnT8d577yEkJAR3797FmjVr5MOD1MXnc+bMGYSFhaFnz544deoUfv31V6xduxZARYgMCgrC7NmzMWPGDDg7O+PixYvYuHEj3nrrrVp5dqlr164YMmQIDhw4AFNTU4XPdcmSJYiIiEDHjh1x7NgxJCQk1Ph8QEWz+dSpUzFz5kxkZWXJe38+CjihoaF45513YGRkhD59+iA3Nxdr166FhoaGPIArw93dHUFBQVi+fDkKCwvh5uaGEydO4Oeff8aiRYtq7dmvkJAQfPfdd5g8eTLee+89GBsbY+vWrbh3716lGRYeDSFjYWGBDRs2wNTUtMo7y5aWlhg5ciQ2btwof1b0l19+gVgsxqJFi9CyZUsIgoAZM2Zg8uTJ0NbWxnfffQcTE5MqQyBRdTG0ET1GJBJhzZo12Lt3L/bv34/jx4+jtLQUjo6OWLBgAYYNGyZ/wLtZs2bYuXMnVq1ahblz50JTUxOdOnVCeHh4tQbj9PT0xHfffYe1a9ciNDQUurq68PX1xcqVK6scCf4RIyMj7Ny5EytXrsSSJUtQWloKFxcXREZGVmsE9w4dOuCbb75BeHg43nnnHVhZWWHs2LHyX2wvUp9IJMK6deuwc+dO/PTTTzhy5AiKi4vRvHlz9OvXD1OmTJE3UT7v/D179sSGDRuwYcMGTJ8+HWZmZhg4cCBmzZr1zJkpavL5vPXWWxCLxZg+fTpatWqF8PBwhQfnP//8c6xbtw5fffUVsrOz0aJFC7z//vuYNGmS0p/788ydOxenT59WWDd8+HDcuHEDO3bswLfffovevXtj/vz5mDNnTo3P17FjR/j4+GDOnDmQSqXo378/5s2bJ9/eq1cvREZGYsOGDdi/fz+MjIzQtWtXfPDBB899Xu5Jjz6/rVu3Ii8vD05OTli1alW1/th4nkfXf8WKFQgLC4NUKkWHDh2wc+fOSn9szJs3DxEREcjJyUFAQAC++OILeVPok+bPnw9zc3Ps3LkTubm5cHFxwddffw0vLy8AFWPdrV69GnPmzEFZWRnat2+PLVu2VPv5RqKqiASBkwASET3i5uaGOXPm1GoAo4Zp//79+PDDD3H+/HmGKlILfKaNiIiISA0wtBERERGpATaPEhEREakB3mkjIiIiUgONuveoTCZDYWEhtLW15T3+iIiIiBoiQRBQVlYGQ0PDKoe/adShrbCwENeuXVN1GURERERKc3V1rXLYmUYd2h5NHu3q6lqtaX2osoSEBIUJp0n98BqqN14/9cdrqP7q+hqWlpbi2rVr8vzypEYd2h41iero6EBXV1fF1ag/fobqj9dQvfH6qT9eQ/VXH9fwaY90sSMCERERkRpgaCMiIiJSA426efRZysrKkJ6ejuLiYlWXoha0tLQgFotVXUa16Onpwd7e/qnPBhAREamTJhva0tPTYWxsjNatW3M4ECUUFhbC0NBQ1WUoTRAEZGdnIz09HY6Ojqouh4iIqMaabPNocXExLC0tGdgaKZFIBEtLS95JJSKiRqPJhjbg6b0zqHHg9SUiosakXkPb4cOH0b9/f7z66qvYuXNnpe2pqakYO3YsXnvtNUyaNAn3798HANy5cwejR49G3759MW3aNBQWFtZn2UREREQqV2+hTSKRIDw8HLt27cLBgwexZ88eJCcny7cLgoBp06Zh8uTJ+N///oe2bdviq6++AgAsXboUo0aNwtGjR+Hp6YnIyMj6KrteLF26FIMGDUL//v3h6emJQYMGYdCgQdi3b1+1jrN7927s3r1b6f1TU1MxdepUBAcHIzg4GO+//z5ycnKe+Zr9+/dj3rx51aqLiIiIaq7eOiKcO3cOAQEBMDMzAwAEBQXh6NGjmDFjBgDgypUrMDAwQGBgIABg6tSpePDgAcrKyhAVFYUNGzYAAEJCQjBmzBjMnj27vkqvc4sXLwZQ0Tli3LhxOHTo0Asd54033lB6X4lEgnHjxiEsLAw9e/aEIAjYtGkTZsyYgV27dr3Q+YmIqLL7BSX49vAVJN2Q4HpOErp4NUMrW2M+wkHVVm+hLTMzE9bW1vJlGxsbxMXFyZf/+ecfWFlZYf78+RCLxXBycsLChQuRm5sLIyMjaGlVlGptbQ2JRFJfZavUjRs3sGjRIuTl5cHAwAAfffQR2rdvj3nz5kFXVxfx8fEoLCzEtGnTMHjwYERERAAAZs6cicOHD2Pjxo0QiUTw8vLCsmXLFIa+2L17NwICAtCzZ08AFc9/TZ48Gfb29igvL0dZWRkWLFiApKQkiEQijBkzBiNGjAAA3Lp1C2PHjsWdO3fQpUsXfPzxxwCAr776Cr/88gukUim6deuG2bNn4/bt23jrrbdgbm4OPT09bN68GStXrsTFixchlUoREhKCCRMm4MKFC9i0aRP09PSQkpICNzc3fP7559DR0cHWrVuxe/duaGpqokePHpg9ezbu3buHRYsWISMjAyKRCO+//z66du1az1eIiOjpBEHA2dg72HQgDoVFZbAx1cLOo1ex8+hVtLA2RBev5uji1QwuLc0Y4Egp9RbaZDKZwjelIAgKy+Xl5bh48SJ27NgBLy8vrF27FsuXL8esWbMqfTNX95s7ISGh0jotLS35s3F/XL6D3y/dqdYxldXDtzkCOzRXat+ioiLIZDJ5Xe+//z4mTJiAXr16IS4uDjNnzsTBgwdRXl6Ou3fv4ttvv0VOTg5Gjx4NHx8flJaWAqgIe59++il27twJW1tbLFiwAMeOHUOPHj3k54qPj0enTp0qPR/Yo0cPlJSUYO3atTAyMsKePXuQm5uLcePGwdXVFSUlJbhz5w52794NfX19vPbaa4iLi0NGRgZiY2Px3XffQSQSYeHChfjhhx/g4+ODGzduICIiAs2bN8eOHTtQVlaGHTt2oLS0FO+88w6cnZ0hk8lw6dIl7N+/H9bW1hg/fjx+++03WFlZYceOHdixYwf09fUxY8YMREVFYdu2bRg4cCC6d++OrKwsTJo0Cbt37640LElpaSliYmJqcgkbFX4W6o3XT33kF0nxc1QurqYXo7mFNkYF2sDWTBsPHkqRdLsI4rQi7Pv9On48eR0mBppoa6+Pti310cpaBxoaDHANmSp/DusttNnZ2SE6Olq+nJWVBRsbG/mytbU1HBwc4OXlBQAYOHAgQkNDYWFhgfz8fEilUmhqalZ6nTI8PT0rzRUmFovlv+B1dXWhqan5om/tmXR1dZUe30xfXx8aGhowNDREYWEh0tPT8dprrwEAunTpAnNzc0gkEmhpaWH48OEwMzODmZkZ/Pz8IBaLoaOjAwBISkqCn58fnJycAADh4eGVzqWtrQ0jI6On1hYTE4NPP/0UhoaGMDQ0RPfu3REfHw8jIyN07NgRzZtXBFEHBwcUFxfj77//xpUrVzB27FgAFUOqtGzZEl27doWlpSVcXFzkxxWLxfJv+ocPH+Kff/5BmzZt4OrqKq/ZxcUFJSUliI+PR69evWBnZwcA2L59OwBgxowZ+Oeff7Bp0yYAgFQqRXZ2dqXvDR0dHXh7eyv1+Td2MTEx8PPzU3UZ9IJ4/dSDIAg4EZWGzUcTUFYmxZsDPTAo0BmamhqIiYlBj5c74dGfz/kPS3HxSgbOx9/FpaRMXLhWAFMjHXRu1wxdvJrB28UK2lp187uJXkxd/xyWlJRUeaPpkXoLbV27dkVERARycnKgr6+P48ePY9myZfLtPj4+yMnJwdWrV+Hu7o6TJ0+iXbt20NbWhr+/P44cOYLg4GAcPHhQ/txbbenp3wo9/VvV6jFrShCEKtdJpVIAUAiZMplM3nwMVNxFfPxu5KPOBRYWFvJ1np6elb4xZDIZQkNDsWTJkkrnf/zcj59LJBLJt40fPx5vvvkmAODBgwfQ1NREbm4u9PT05PtLpVLMnj0br776qrw2Q0NDXL58WSFYPzruk+9FIpFAX18fMpkM3333nfwZyczMTFhaWlb+IImI6klm7kNs+CEWl5Iy4eFogdARPmhhbfTU/Y0NdNCrYyv06tgKRSXliLkqwfm4uzhzOR3HL9yCgZ4WOra1Q5f2zeDnZgM93SY7Hj79q956j9ra2mLWrFkYN24cBg8ejIEDB6J9+/aYPHky4uPjoaenhw0bNmDBggUYMGAALly4IO+luHjxYuzduxf9+/dHdHQ03nvvvfoqW2WMjIxgb2+P48ePAwAuX76Me/fuye9Y/fLLLxAEAbdv30ZcXJxC8vfy8sLly5eRlZUFAPj0009x4sQJheOPGDECp0+fxunTpwFUhLLIyEhkZ2fDysoKAQEB+PHHHwFUBKtTp06hU6dOT603ICAAhw4dQmFhIcrLy/HOO+/g2LFjVe63d+9elJWVobCwEKNGjcLly5efelx/f3+cPn1aftz3338fCQkJCAgIkHeYSE5ORnBwMIqKip73sRIR1TqZTMCRczcwY9VJJN7IxpQhXvhserdnBrYn6etqoZt3C8we648dS/th0aTOeKl9c1xKysTy76IwevFRfLr1In6PSUNBUVkdvhtqyOo1tj8aWuJxX3/9tfxrb29veVB4XIsWLeTNYk3JqlWrsGTJEkREREBbWxsRERHyJtDi4mIMHToUpaWlCAsLg7m5ufx1tra2+OijjzBp0iTIZDJ06NABISEhCse2trbG119/jZUrV+Lzzz+HVCqFh4eHvJfuO++8gyVLliA4OBhSqRSTJk1Cu3btkJSUVGWtPXv2xNWrV/H6669DKpXi5ZdfxpAhQ3D79m2F/UaOHIlbt25hyJAhKC8vR0hICDp37owLFy5Uedx27dphzJgxGDlyJGQyGfr06YOuXbvC2dkZixYtkn8/rVy5EkZGyv8DSaQuZDIBaZJ8JN8pRocOMmhqNukx0RucO/cKELH3MhJSstHBxRozXu8AWwuDGh1TR1sTHT3s0NHDDu8Mk+HKjWycj7uLc/F3cT7+LjQ1RPB2sUYXr2bo7GkHc2O95x+UGgWRUFU7XCPxqG34ac+0tW3bVkWV1cy8efPQqVOnSkGsLqnb3KOPqPN1rm18Jko9lJXLkHI7D4mp2biSmgPxzWzkP6y4s9LC2gjjB7RFgGcz9jZUMalMwP/+SMGOo1ehrSnCxNc80adTq2del5r+DMpkAq6l5eJ8XEV4u5tdCJEI8HC0RBevZuji2Qw2NQyM9Gz19UxbVbkFaMITxhMRNQRFJeVIupWDK6k5SLyRjau3clFaVvH8aHMrQwR4NoOHoyVup9/CX9dL8enWKLg5mGPCAA94OlupuPqm6VbGA0TsuYykf3LRycMO04e1h6Wpfp2fV0NDBHcHC7g7WGDCQA/cvPsA5/+9+7b5UAI2H0pAG3tT+VAiLW2N67wmql8MbWpo+fLlqi6BiF7Q/YISJN6oCGhXUrORcvs+ZDIBGiKgdXNTBAU4oJ2jJTwcLWBu8l+zV4xmFsYMegknotOw8+hVfBj5J/zb2mL8AA+0bmaiwnfUdJRLZdh38jq+//Ua9HW18MFoPwT6tFDJXU+RSATH5qZwbG6KUUHuuJNVIA9w238RY/svYrS0NZIHOOcWprw72wgwtBER1aHMnIe48m9AS7yRjTRJAQBAW0sDrq3MMbRHG7RzsoS7gwUM9bWfeSxNTQ282tkBgT4t8NPZG/jxxDWErv4dPfxaYnRfd9iYs2msrqSk5+GLPZeReuc+Xu7QAm8P9oKZceXmK1Vpbm2EoT1dMLSnC+7lFeGvhIoA9+OJa9j72zXYmOvLA1zb1hYcC05NNenQ9uQAv9S4NOLHNamBkskEpGXmy59Hu3IjG/fyKno1G+hpoW1rC/TwawkPR0u4tDSDjvaLjcGlp6OFYT1dEBTggB9OXMdPZ1Nx5vJtDHjJEcN7ucLEUKc231aTVlomxfe/JmHf78kwNdTB/Amd0MWrmarLeiYrM30M7OaEgd2ccL+gBBevZOBc/F38/OcNHPojBebGugjwrBgLzquNFbTYuUVtNNnQpqenh+zsbFhaWjK4NUKCICA7O1thjDii2lYulSElPU/+PFrijf86DZgb68LDyRIhr1TcSXNoZgLNWr67YWygg4nB7TCwmyN2HbuKQ3+k4NcLtzC0pwuCX3aCnk6T/Se+Vly9lYMv9vyNNEkBenVsibde84SRgXoFYlMjXfTp7IA+nR3wsLgM0WIJzsXfxe8xafjl/E0Y6mujczs7dPFqBh83G+i+4B8SVD+a7E+0vb090tPT5WOZ0bOVlpbKhxtRF3p6erC3t1d1GdSIFJeUI+lWrry5M+mfXJSUVnQaaGZliM7tmqGdkwU8nCzRzNKw3v4gtDE3wHsjfTGkextsOyLGtiNi/HT2BkYFuaF3x1YcJqSaikvLseOXq/jfmRRYmupj6eQu8HWv3kw8DZGBnjYCfewR6GOPkjIpLidl4lz8XVy8koGT0WnQ1dGEv7stung1Q0cPWxjoPbu5nupfkw1t2tracHR0VHUZaiMmJobTQVGT86CwVN5hIPFGNlLS70MqEyASAY7NTNGnUyu0c7KEh6MlLExUf1fXoZkJFk7qjCup2dj60xWs/yEWB06lcJiQaohLzkLE3svIyH6Ifl1bY8IAj0YZXnS1NdHZsxk6ezZDuVSGhJR7OBd/F3/F38WfcXegpamBDq7/jgXXzg6mRg3n+b2mrMmGNiKiJ2XmPqx4Hu1GDq6kZiNNkg8A0NLUgGsrM4T0aAMPR0u0bf38TgOq1M7JEitnvoy/EjKw7UgihwlRwsPiMmz9KRG/nL+JZpaG+HT6S/BqIp9VRUCzQQdXG0wd0h5Xb+XgfHzFYL7RYgk2iIB2TlYVY8F5NYOVWd0Pb0JVY2gjoiZJECpmGrhyI+ffoJaNrNyKTgP6ulpo62iBV3zt0c6pZp0GVEUkEqGLVzN08rDFb1Fp2HWsYpiQjh62GN/fAw4cJkQuWizBhh9jkXO/CIO7O2N0X/cm+zyghoYIHo4Vd48nBrdD6u378gD31cF4fHUwHq6tzNDFqzm6ejVD82pM1UU11zS/K4moySmXypB6+z6upD5q7sxB/sNSAICZsS7aOVpicHdntHO0ROvmprXeaUBVNDU1EBTggO6+/w0TMpPDhAAA8h+WYvOhBJyMTkNLW2OsnPky3BwsVF1WgyESieBsbwZnezOM6dcWaZJ8/JVQEeC++zkR3/2cCAc744oA174ZWjczYRN8HWNoI6JGqbi0otPAo7toV2891mnA0hCd2tminaMl2jlZoplV/XUaUBUOE6LofPwdRO6LQ35hKUb0dsWIPq7Q1lKvu6n1raWtMVraGmN4L1dk5jyUB7g9vyXh+1+TYGdpAB9XG3i7WMOrjVWT+56qDwxtRNQo5D8slT+PlpiajeT0PHmngdbNTNCnYyt4OFXMNFAfUw41VE19mJC8/BJ8eSAOf8begVMLUyyd3AVOLUxVXZbasbEwwGuBzngt0Bm5+cW4eCUDfyVk4NSliqFERCLAsbkpvF2s4e1ihXaOltDTbdzfW/WBnyARNWhSqQx5BSXIfVCCnPxi5D4oRs6Dkn//X4zc/IrlR4PYamlqwKWlGYb8Oz6ae2sLGDXgTgOq8vRhQtzRu2PLRjdMiCAIOH0pHV8dTEBRSTnG9W+LIa+04cCytcDcWA9BAa0RFNAa5VIZrv+Th9jkLMRez8LhMyk4cCoZWpoiuDlYwLuNFdq7WMPNwZyf/QtgaCMilSgrlyqGrwfFyMl/fLkipD0oKIGsisktjA10YGGiC3MTPdjbGKO5tSHaOVrCtZW52nUaUKXKw4RcxsHTyRjX3wMBnnaNotk4+34RNvwYi6hECdwczPHuCB9Opl5HtDQ10NbRAm0dLTCyjxuKS8uReCMHcdcrQtzuX5Ow63gS9HQ04elsBW8XK3i7WMPBzoRTaymBoY2IatXD4jLk5pf8F8QeBbMn7pIVFJVVeq2GqKJTgLmJHixM9dCmpRnMTXRhYaIHc2M9eUgzN9aDthb/Sq9NlYcJuQh3B3NMGNgO7ZwsVV3eCxEEAccv/INvDyegXCrgrUGeGNjNqdF0MlEHejpa8HWzga9bxeDE+Q9LEZ98D5evZyHuehaixRIAgImhDtq3sfq3OdUadpYGjeIPhtrG0EZEzyUIAgqKyp4SxB4PaMUo/vdh/8dpaWo8dlfMCF7OFYPRmpvo/RvIKoKZiZEuf6GqUFXDhMzbcFYthwnJyC7E+h8uI/b6PXg5W2Hm6x3QzMpQ1WU1ecYGOujavjm6tm8OALiXV4TYf+/CxV6/h7OxdwAANub68HaxRnsXa3i3sYJ5Axi8uiFgaCNqwqQyAQ8K/g1d+Yrh6/Hl3PwSlJXLKr1eX1cT5sYV4cvZ3gz+JrqwMH4UxnTlocxIX5t/NauRx4cJOXwmFftOXsfM1b+jp39LjApq2MOEyGQCfv7zBr47kggNkQjTh3kjqLMDm94aKCszffTq2Aq9OraCIAhIzyyoaEpNrpih4deL/wAAWtkZV9yFa2MFT2erBj24dV1iaCNqxKRSGW7efYDk9PuIE9/HuZTL8of3cx8UI6+gFLIqHhgz0teGhakeLIz10Nzpsbtixnr/NVea6EGfvcEaNT0dLQzv5Yq+XVrLhwn54+/bGNjNCcN7ucC4gU2enp6Zjy/2XIb4Zg783G3wzrAOsDZvuj2F1Y1IJJIPKzKgmxOkMgGpt/MQe/0eYq9n4dhft3D4TCo0NERwsTdD+3+fh2vb2qLJPMfKf3GJGglBEJCVV4Rr/+Qi6VYurv2Ti+T0+ygt+6+50sy4VB68nJqb/hvE/rsjZv5vU2VT+QeQlPPkMCEHTyfj+F83G8wwIVKpDAdOp2DXsavQ1dbErDd80MOvJe/uqjlNDRFcWprDpaU5hvV0QVm5FFdv5sqbU/f9nowfTlyHjlZF54dHz8M525s12scsGNqI1NTD4jJcT8tTCGm5+SUAAG0tDTi1MEXfAAe4tjKHaytzpN0Qo1NHfxVXTers0TAhg7u3wbYjiQ1imJCbdx9g3feXkJx+H128mmFaSHs+/9RIaWtpwquNFbzaWGFMv7Z4WFyGhNRsxF7PQtz1e9h2RAxADEN9bXg5W8pDnL2NUaMJ8AxtRGpAKhPwT8YDeUBL+icXaZJ8CP+2bDa3MoS3qzXc/g1ojs1NK/WuvHOrcfyjRarXupkJFk0KUOkwIWXlMvxw4hp+OHENRvo6mDeuI17ybl7n56WGw0BPG5087NDJww4AkJtfjPjke/Lm1L8SMgAAFia6/3ZoqAhx6txkztBG1ABl3y+S3z1L+icXyWl58l6ZxgbacG1ljm7tm8PVoSKkNbRni6hpUNUwIdfTcvHFnsu4efcBXvGzx+RBXpwyiWBurIdAH3sE+tgDqOhB/KhX6t9JmTgVkw7g3z9y/70Lp27TbTG0EalYcUk5rqfn4dq/d9Cu/ZOL7PvFAAAtTREcm5uid8dWcHUwh1sr8yYxTyapj6cNE9LJww7j+ret1WFCSsqk2H3sKg6cSoa5iR4WTuosv8tC9CQ7S0PYWRoiKKA1ZDIBtzIeyO/Cqet0Ww23MqJGSCoTkC7Jl4ezpFu5+CfjgXzEfztLA7Rzsqxo5nQwh1NzU3YKILVQ1TAhoat/R49aGibkSmo2Ivb+jdtZhQgKcMCbA9s12WEfqPo0NCr+AHZsborB3Z2rmG4rVXG6rX9DnGurhjXdFkMbUR3KfVCsENCup+WhqKQcAGCorw3Xlmbo7OkqfxbN1EhXxRUT1cyjYUKCAlrjx5M1HyakqKQc235OxM/nbsDa3AAfT+kKb1frOqqemornTrd1/Cp2HasYi7Kd03/TbQlCFXPq1WfdKj07USNSXFqOlPT78ufQrv2Ti6zciknMNTVEcGxugh5+9nD79zm05lZGHPCTGi0Tw6qHCRnWyxXBLztBV4k7yJevZSLih1hk5T7EwG5OGNuvLccGpDrxtOm2Hj0T92i6rQEdzeCvwk74/O4negEymYDbWQUKnQVu3n0gH6jWxlwfbq3M8drLTnBrZQEne1OlfkkRNTZPDhPy3c+J+OlsKkYFuaOXf9XDhBQUleHb/yXg14v/oIW1IZa/0w0ejuo5/ympp6qm2xLfzEF5/m2V1sXQRqSE+wUlSHo0HtqtXFxPy0VhcUUzp4GeFlxammFojzbyZ9HMjTlOFNHjHg0TkpByD1t/TkTE3ophQsb2Uxwm5OKVDGz4MRZ5BSUY2qMN3ghy5x88pHJWZvp4uUMLxMRkqLQOhjaiJ5SWSZF6+35FE+e/PTolOQ8BVDzM2trOBC/72MOtlRlcW5nD3saYzZxESvJ0tsKqKoYJGdHHDadi0nH673S0bmaChRM7o01LM1WXS9SgMLRRk5d9vwix1+/918x55z7KpRXNnFZmFc2c/bs6ws3BHM4tTBt0d3AidVDVMCFLN/8FLU0RRgW5Y1hPl0qDQxMRQxs1QYIg4ObdB7hwJQMXr2TgeloegIpeQm3szTEo0FneWcDSVH1HziZq6B4fJuRc3F0425vCwa72xnUjamwY2qhJKCuXIj4lGxevZODClQzcyyuCSAS4tjLH2H5t4d/WFg7NTBrtJMNEDZmejhZ6+rdUdRlEDR5DGzVaDwpLES2W4OKVDFxKkqCoRAodbU34uFpj1Ktu8PewZYcBIiJSG/Ua2g4fPoyNGzeivLwc48ePx+jRoxW2r1+/Hvv27YOJScXt8ddffx2jR4/GgQMHsHr1alhaVnT5fuWVVzBr1qz6LJ3URHpmPi5eycDFRAnEN7IhEyomCw70sUfndnZo72LNnmhERKSW6i20SSQShIeHY//+/dDR0cHIkSPRuXNntGnTRr5PQkIC1qxZAx8fH4XXJiQkYN68eRg4cGB9lUtqQiqVQXwzR/582p17hQAAp+amGN7bFZ3b2cG5hRl7dxIRkdqrt9B27tw5BAQEwMzMDAAQFBSEo0ePYsaMGfJ9EhISsGnTJty+fRsdO3bE3Llzoauri/j4eNy8eRObNm2Cm5sbFi5cCFNT0/oqnRqYh8VluJSUiQtXMhAjliD/YRm0NEVo38YarwU6o6OHbY3nOSQiImpo6i20ZWZmwtr6v/nibGxsEBcXJ18uLCxE27ZtMXv2bDg4OGDevHmIjIzErFmzYG1tjYkTJ8LX1xdr1qxBWFgYVq9eXV+lUwMgyXn4b7NnBhJS7qFcKsDYQAcdPezQqZ0dfFytYaDHyaOJiKjxEgn1NPvpxo0bUVJSgvfeew8AsHfvXiQkJCAsLKzK/RMTEzF//nwcPHhQYf39+/fRp08fXLx48bnnLCkpQUJCQk1LJxWQCQLuZJch6XYRrt0uhiSvDABgZaIF1xZ6cGuhj5ZWOmz2JCKiRsfT0xO6urqV1tfbnTY7OztER0fLl7OysmBjYyNfvnPnDs6dO4dhw4YBqBhLS0tLC/n5+di3bx8mTJggX6+pWb0HyZ/25kl5MTEx8PPzq9NzFJeWI+76PVy4koGoxAzk5pdAQ0MED0cLDHi54o5aC2ujOq2hMauPa0h1h9dP/fEaqr+6vobPu9lUb6Gta9euiIiIQE5ODvT19XH8+HEsW7ZMvl1PTw+rVq1C586dYW9vj507d6JPnz4wMDDA5s2b4ePjA29vb+zYsQN9+vSpr7KpjuU+KMbFxIphOS5fz0JpmRT6ulrwc7dB53Z28GtrC2MDHVWXSUREpHL1FtpsbW0xa9YsjBs3DmVlZRg2bBjat2+PyZMnIzQ0FF5eXggLC8O0adNQVlYGX19fvPnmm9DU1MTatWuxZMkSFBcXo3Xr1li5cmV9lU217NFsBI+eT7v2Tx4AwMZcH692boXO7ezQzsmKU9gQERE9oV7HaQsODkZwcLDCuq+//lr+dVBQEIKCgiq9zt/fHwcOHKjz+qhulJXLkJByTx7UMnOLAABurcwxpp87OrdrBgc7Y4hEfD6NiIjoaTgjAtWJB4WliLkqwYUrGbh0NRNFJeXy2QhG9HFDx7a2MDfhbARERETKYmijWnM7q0A+t6fibAQt0KmdHdq3sYKeDr/liIiIXgR/g9ILk0pluHorVx7UbmcVAAAcm5tgeC9XdGpnhzb2nI2AiIioNjC0UbU8LC7D30lZuHDlLqLFmch/WAotTRG8nK0wsJsjOnnYwcaCsxEQERHVNoY2eq77BSW4eK0Ah6LPIT4lG+VSGYwNtOHf1had2zWDjxtnIyAiIqprDG30XIs2nUfqnftoYW2E4Jed0LmdHdwdzKGpyWE5iIiI6gtDGz3TnXsFSL1zH306mCJ07CuqLoeIiKjJ4q0SeqboRAkAoG0rfRVXQkRE1LQxtNEzRYklsLcxgoURb8oSERGpEkMbPVVRSTkSUrLh39ZW1aUQERE1eQxt9FSXr2WhXCpDRw+GNiIiIlVjaKOnihZLYKCnBQ9HS1WXQkRE1OQxtFGVBEFAtFgCH1cbaHFoDyIiIpXjb2OqUurt+8h5UMzn2YiIiBoIhjaqUrS4YqgPv7Y2Kq6EiIiIAIY2eooosQQuLc1gbqyn6lKIiIgIDG1UhfsFJbj2Ty46smmUiIiowWBoo0pirmZCEAB/DvVBRETUYDC0USXRYgnMjHXh3MJM1aUQERHRvxjaSIFUKsOlpEz4u9tCQ0Ok6nKIiIjoXwxtpEB8MweFRWVsGiUiImpgqh3aysvL66IOaiCixRJoaYrg42qt6lKIiIjoMUqHtp07d6JPnz7o0KED0tLSsHDhQqxZswYymawu66N6Fi2WwMPREgZ62qouhYiIiB6jVGjbsmULNm/ejHfeeQdaWloAgJdeegn79u3D2rVr67I+qkeZuQ9xKyOfE8QTERE1QEqFtu+//x5hYWEYPHgwRKKKh9P79u2LlStX4uDBg3VZH9WjR7MgcOoqIiKihkep0JaRkYHWrVtXWm9nZ4cHDx7Udk2kIlGJEjSzNEQLayNVl0JERERPUCq0dejQAfv27VNYJ5PJsHnzZrRv375OCqP6VVImRVzyPfh72MrvphIREVHDoaXMTgsWLMDkyZNx6tQplJaWYvHixbh58yZKS0uxefPmuq6R6kF88j2UlknZNEpERNRAKRXanJ2dcezYMRw+fBgpKSmQSqXo06cPXnvtNRgaGtZ1jVQPohIzoKejCS9nS1WXQkRERFVQKrS99tpr+PzzzzFs2LC6rodUQBAERIsl8HaxhraWpqrLISIioioo9Uzb/fv3oanJX+aN1T+SfGTmFnGoDyIiogZM6TttEydORHBwMFq0aAFdXV2F7bwDp96iEznUBxERUUOnVGj75ZdfoK2tjaNHj1baJhKJGNrUXJRYAqfmprA01Vd1KURERPQUSoW2kydP1nUdpCIFD0shvpmDYT1dVF0KERERPYNSoQ2oGGB3+/btSElJgUwmg5OTE4YPHw5nZ2elT3b48GFs3LgR5eXlGD9+PEaPHq2wff369di3bx9MTEwAAK+//jpGjx6NO3fuYPbs2cjOzoajoyM+//xz9lqtJX8nZUEmE9CRTaNEREQNmlIdES5evIi+ffsiJiYGrVu3hoODAy5duoQhQ4YgJiZGqRNJJBKEh4dj165dOHjwIPbs2YPk5GSFfRISErBmzRocOnQIhw4dkoe6pUuXYtSoUTh69Cg8PT0RGRlZzbdJTxMlzoCJoQ5cWpmruhQiIiJ6BqVC24oVKzBu3Dh8//33mDdvHj766CPs3bsX48ePx6pVq5Q60blz5xAQEAAzMzMYGBggKCio0jNyCQkJ2LRpE4KDgxEWFoaSkhKUlZUhKioKQUFBAICQkJAqn62j6pPKBMRczYSvuw00NTgLAhERUUOmVPNocnIy1qxZU2n90KFDsW3bNqVOlJmZCWtra/myjY0N4uLi5MuFhYVo27YtZs+eDQcHB8ybNw+RkZEYPXo0jIyMoKVVUaq1tTUkEolS53wkISGhWvs3FWn3SvCgsBSWug+VumOq7F1Varh4DdUbr5/64zVUf6q8hkqFNnt7e8TGxsLBwUFh/eXLl2FpqdwI+jKZTGFOS0EQFJYNDQ3x9ddfy5cnTpyI+fPnY9SoUZXmwqzu3Jienp6VhikhQPyLGBoa9zCsX2cYGeg8c9+YmBj4+fnVU2VUF3gN1Ruvn/rjNVR/dX0NS0pKnnmjSanQNmnSJCxevBjJycnyCeJjY2Oxc+dOvP/++0oVYmdnh+joaPlyVlYWbGxs5Mt37tzBuXPn5MOHCIIALS0tWFhYID8/H1KpFJqampVeRy8uSixB29YWzw1sREREpHpKPdMWEhKChQsX4uzZs3j//ffx0UcfISoqCsuXL6/UA/RpunbtivPnzyMnJwdFRUU4fvw4AgMD5dv19PSwatUqpKWlQRAE7Ny5E3369IG2tjb8/f1x5MgRAMDBgwcVXkcvJvt+EVJv3+eAukRERGpC6SE/QkJC8Oqrr8LIyAhAxXNubdq0UfpEtra2mDVrFsaNG4eysjIMGzYM7du3x+TJkxEaGgovLy+EhYVh2rRpKCsrg6+vL958800AwOLFizFv3jxs3LgRzZo1q/L5OqqeaHEmAHCoDyIiIjWhVGhLTU3FO++8gx49emDOnDkAgAkTJsDc3ByRkZFo2bKlUicLDg5GcHCwwrrHn2MLCgqS9xJ9XIsWLbB9+3alzkHKiRZnwNpcH63sjFVdChERESlBqebRsLAwtG/fHtOnT5evO378ONzd3bF06dI6K47qRlm5FJevZcG/rW21O3UQERGRaigV2mJjYzFjxgx50ygAGBgYYMaMGbh06VKdFUd1IyElG8WlUjaNEhERqRGlQpulpSXi4+MrrU9KSpJPOUXqI1osgY6WBrzaWKm6FCIiIlKSUs+0jR8/HosWLUJycjI8PDwAAGKxGNu3b8eUKVPqtECqfVFiCdq7WENPR+l+KERERKRiSv3WHjt2LPT09LBnzx5s2bIF2tracHBwwMKFCyt1LKCG7XZWAe7eK8Sgl51UXQoRERFVg9K3WoYPH47hw4fXZS1UD6ISK6YA8/ewU3ElREREVB3PDG1FRUX4/fff0b17dxgaGgIAtm/fjj///BMWFhYYN24c3N3d66VQqh3R4gy0tDWGrYWBqkshIiKianhqR4S0tDT069cPH330EbKzswEAn332GT799FPo6OhAQ0MDo0ePRmxsbL0VSzXzsLgMV1Kz2WuUiIhIDT31TtvatWvh7u6O8PBw6OvrIzs7Gzt27EDfvn0RHh4OAHBycsLatWuxZcuWeiuYXtzla1kolwrw92BoIyIiUjdPvdN27tw5zJgxA/r6+gCAP/74AzKZDIMHD5bv89JLLyEuLq7Oi6TaES2WwFBPC21bW6i6FCIiIqqmp4a2wsJCWFj898v9/Pnz0NHRQUBAgHydnp4eZDJZ3VZItUImExAtlsDHzQZamkoNz0dEREQNyFN/ezs4OODq1asAgNLSUvzxxx/o3LkzdHV15fucOXMGDg4OdV8l1Vjq7fvIzS9BRzaNEhERqaWnPtM2cuRILF26FLdu3UJUVBTy8vLw5ptvAvivV+m6deswY8aMeiuWXlyUWAKRCPBzZ2gjIiJSR08NbaNHjwYA7Nu3DyKRCCtWrECXLl0AACtXrsTevXvxxhtvYNy4cfVTKdVItDgDri3NYWqk+/ydiYiIqMF55jhto0ePloe3x02ZMgUzZ85UeOaNGq68/BJcT8vDqCCOqUdERKSuXmjySTs7jqavTmKuSiAIgD/HZyMiIlJb7EbYBESJJbAw0YVzC1NVl0JEREQviKGtkSuXyvB3Uib83G0hEolUXQ4RERG9IIa2Rk58IwcPi8s51AcREZGaUzq03bt3Dxs3bsTcuXORnZ2NI0eOyMdxo4YrSiyBlqYI3i7Wqi6FiIiIakCp0BYfH4+goCCcP38eP//8Mx4+fIiLFy/i9ddfx9mzZ+u6RqqBaHEGPJ2sYKCnrepSiIiIqAaUCm3Lly/H22+/jW3btkFbu+KX/5IlS/D2229j9erVdVogvbiM7EKkSQo4QTwREVEjoFRoS0xMRL9+/SqtHzRoEFJTU2u9KKod0WIJAKAjh/ogIiJSe0qFNktLS6SkpFRaHxMTAxsbm1ovimpHlFiC5laGaG5tpOpSiIiIqIaUGlx38uTJWLhwISZPngxBEPDnn3/i7t272LZtGz744IO6rpFeQHFJOeKT76Ff19aqLoWIiIhqgVKhbcSIEbC2tsY333wDPT09rF69Go6Ojvjkk0/Qv3//uq6RXkBc8j2UlcvYNEpERNRIKD2NVc+ePdGpUycYGVU0tSUnJ6NNmzZ1VhjVTJRYAn1dTbRzslJ1KURERFQLlHqmLTU1Ff369UNkZKR83YQJExAcHIy0tLQ6K45ejCAIiE7MQAdXG2hrcfxkIiKixkCp3+hhYWFo3749pk+fLl93/PhxuLu7Y+nSpXVWHL2Ym3cf4N79Yk4QT0RE1IgoFdpiY2MxY8YMedMoABgYGGDGjBm4dOlSnRVHL+bRUB8MbURERI2H0kN+xMfHV1qflJQEExOTWi+KaiYqUQJne1NYmOipuhQiIiKqJUp1RBg/fjwWLVqE5ORkeHh4AADEYjG2b9+OKVOm1GmBVD0PCkuRdCsHw3u7qroUIiIiqkVKhbaxY8dCT08Pe/bswZYtW6CtrQ0HBwcsXLgQwcHBdV0jVcOlpEzIBM6CQERE1NgoPeTH8OHDMXz48LqshWpBdKIEpkY6cGlprupSiIiIqBYpHdrOnDmD+Ph4lJeXQxAEhW3vvvuuUsc4fPgwNm7ciPLycowfPx6jR4+ucr9Tp04hLCwMJ0+eBAAcOHAAq1evhqWlJQDglVdewaxZs5QtvcmQygRcSpKgo4cdNDREqi6HiIiIapFSoe3TTz/Fjh074O7uDkNDQ4VtIpFy4UAikSA8PBz79++Hjo4ORo4cic6dO1caoPfevXtYsWKFwrqEhATMmzcPAwcOVOpcTVXSrRzkPyxjr1EiIqJGSKnQduTIESxbtgxDhw594ROdO3cOAQEBMDMzAwAEBQXh6NGjmDFjhsJ+CxYswIwZM7B69Wr5uvj4eNy8eRObNm2Cm5sbFi5cCFNT0xeupbGKFkugoSGCj5uNqkshIiKiWqZUaJNKpfD19a3RiTIzM2FtbS1ftrGxQVxcnMI+27Ztg4eHB7y9vRXWW1tbY+LEifD19cWaNWsQFhamEOqeJyEhoUa1q4s/YiRoaaWNpMS45+/8AmJiYurkuFR/eA3VG6+f+uM1VH+qvIZKhbYxY8Zg/fr1CAsLq9Q8qiyZTKbQlCoIgsLytWvXcPz4cWzduhUZGRkKr92wYYP867feegt9+vSp1rk9PT2hq6v7QnWri6zcIkjy0vHmQA/4+bnU+vFjYmLg5+dX68el+sNrqN54/dQfr6H6q+trWFJS8swbTUqFtj///BNxcXH45ZdfYG5uDm1tbYXtp06deu4x7OzsEB0dLV/OysqCjc1/zXhHjx5FVlYWhg4dirKyMmRmZmLUqFHYtGkT9u3bhwkTJgCoCHuamprKlN2kRF/lLAhERESNmVKhrTaG++jatSsiIiKQk5MDfX19HD9+HMuWLZNvDw0NRWhoKAAgPT0d48aNw65duyCVSrF582b4+PjA29sbO3bsqPadtqYgOlECGwsDtLQ1VnUpREREVAeUCm1DhgwBUNHEefv2bTRr1gwymQw6OjpKn8jW1hazZs3CuHHjUFZWhmHDhqF9+/aYPHkyQkND4eXlVeXrNDU1sXbtWixZsgTFxcVo3bo1Vq5cqfR5m4LSMilik7PQu2MrpXvzEhERkXpRKrSVl5dj9erV2LFjB6RSKY4dO4bPP/8cWlpaWLZsGQwMDJQ6WXBwcKUZFL7++utK+9nb28vHaAMAf39/HDhwQKlzNEXxKfdQUipl0ygREVEjptSE8evWrcPZs2fx7bffyh/oHzt2LBITE7F8+fI6LZCeLzpRAh1tTXi1sVJ1KURERFRHlAptP//8M5YsWYKOHTvK1/n7++PTTz/Fr7/+WmfF0fMJgoAosQTeLlbQ1WYHDSIiosZKqdCWm5srn0Lqcfr6+iguLq71okh56ZkFkOQ85ATxREREjZxSoa1Lly74+uuvFeYczc/Px5o1axAQEFBnxdHzRSVWDPXhx9BGRETUqCkV2hYvXoykpCR06dIFJSUlmDJlCgIDA5GRkYEFCxbUdY30DNFiCVo3M4GNuXKdQYiIiEg9KdV7VFdXFz/++CPOnz+P1NRUlJeXw9HREd26dYOGhlK5j+pAYVEZEm9kY8grbVRdChEREdUxpULboEGDEBkZiS5duqBLly51XRMp6e9rmZDKBA71QURE1AQodZtMV1cXpaWldV0LVVNUogRG+tpwdzBXdSlERERUx5S60/byyy9j4sSJCAwMRIsWLSpNvv7uu+/WSXH0dDKZgEtXM+HrbgNNTTZRExERNXZKhbZr167B09MTOTk5yMnJUdjGaZNUIzk9D3kFJRzqg4iIqIlQKrRt3769ruugaopKlEBDBPi6M7QRERE1BUq3q129ehULFizA2LFjIZFIsGPHDvzxxx91WRs9Q7Q4A24OFjAx1FF1KURERFQPlAptZ86cwciRIyGVShEbG4vS0lLk5ORg+vTpOHz4cF3XSE/IeVCM5PT77DVKRETUhCgV2sLDw/Hhhx/is88+g6ZmxfyWoaGh+OijjxAZGVmnBVJlMeKKWRA6ejC0ERERNRVKhbbU1FR07dq10vqXXnoJt2/frvWi6NmixBJYmeqhdTMTVZdCRERE9USp0GZvb4/Lly9XWn/y5Em0bNmytmuiZygrl+HytSz4tbVlz10iIqImRKneo++99x7mzJmD+Ph4SKVS7Nu3D2lpaTh27BhWrVpV1zXSYxJTs1FUUs6hPoiIiJoYpe609e7dG7t27UJeXh5cXFxw6tQpyGQy7Ny5E/369avrGukxUWIJtLU04O1irepSiIiIqB4pdacNANzd3bFy5cq6rIWUEC3OgJezFfR0lb50RERE1Ago9Zu/tLQUP/zwA5KSklBSUgJBEBS2M8zVjzv3CnA7qxADXnJSdSlERERUz5QKbQsXLsTx48fx0ksvwdjYuK5roqeITuRQH0RERE2VUqHt+PHjiIyMRJcuXeq6HnqGKLEE9jZGsLM0VHUpREREVM+U6ohgYmICa2s++K5KRSXlSEjJ5iwIRERETdRTQ5tMJpP/N3XqVCxbtgypqakoKytT2CaTyeqz3ibr8rUslEtlbBolIiJqop7aPOrh4aEweKsgCBgwYECV+4rF4tqvjBREiyUw0NOCh6OlqkshIiIiFXhqaNu2bVt91kHPIAgCosUS+LjaQEtTqRZtIiIiamSeGto6depUaV1KSgpSUlIglUrh6OgId3f3Oi2OKqTevo+cB8V8no2IiKgJU6r36P379zF37lycOnUKpqamkEqlKCwshL+/PyIjIzkMSB2LFlcM9eHX1kbFlRAREZGqKNXWtmzZMmRlZeGXX37BhQsXEB0djcOHD6OoqAifffZZXdfY5EWJJXBpaQZzYz1Vl0JEREQqolRo+/3337F06VI4OjrK17Vp0waLFi3CiRMn6qw4Au4XlODaP7mcIJ6IiKiJUyq06elVfYdHJBJBKpXWakGkKOZqJgQB8OdQH0RERE2aUqGtZ8+eCAsLw40bN+TrUlNTsWzZMvTo0aPOiqOK59nMjHXh3MJM1aUQERGRCinVEWH27Nl455130K9fPxgZGQEACgsL0b17dyxcuLBOC2zKpFIZLiVlootnM2hoiJ7/AiIiImq0lAptJiYm2L59O5KSkpCSkgJdXV04OTkpPONGtU98MweFRWVsGiUiIqLnN4/GxcWhpKQEAODm5ob+/fsDqBgGpLoOHz6M/v3749VXX8XOnTufut+pU6fQs2dP+fKdO3cwevRo9O3bF9OmTUNhYWG1z62OosUSaGmK4OPKeV+JiIiauqeGtvLycsyePRsjRoxAbGyswraffvoJb7zxBhYsWKB0RwSJRILw8HDs2rULBw8exJ49e5CcnFxpv3v37mHFihUK65YuXYpRo0bh6NGj8PT0RGRkpFLnVHdRYgk8HC1hoKet6lKIiIhIxZ4a2r799ltcuHAB27ZtqzQ7Qnh4OLZs2YITJ05g+/btSp3o3LlzCAgIgJmZGQwMDBAUFISjR49W2m/BggWYMWOGfLmsrAxRUVEICgoCAISEhFT5usYmM+ch/snI5wTxREREBOAZz7QdOHAACxcuRMeOHavcHhAQgDlz5uCbb77BhAkTnnuizMxMWFv/18xnY2ODuLg4hX22bdsGDw8PeHt7y9fl5ubCyMgIWloVpVpbW0MikTz3fI9LSEio1v4NwcVrBQAAfdk9xMRUvym6LsTExKi6BKohXkP1xuun/ngN1Z8qr+FTQ9vdu3fh4eHxzBf7+/tj6dKlSp1IJpNBJPqvB6QgCArL165dw/Hjx7F161ZkZGQ8dT8AlZafx9PTE7q6utV6jar99PdfaGZpiFdf6Vzt91sXYmJi4Ofnp+oyqAZ4DdUbr5/64zVUf3V9DUtKSp55o+mpzaNWVlZIT09/5sHv3LkDc3NzpQqxs7NDVlaWfDkrKws2Nv/NpXn06FFkZWVh6NChePvtt5GZmYlRo0bBwsIC+fn58mfnnnxdY1RcWo6461nw97BtEIGNiIiIVO+poa1Pnz6IiIhAWVlZldvLysqwfv16BAYGKnWirl274vz588jJyUFRURGOHz+u8NrQ0FAcO3YMhw4dwldffQUbGxvs2rUL2tra8Pf3x5EjRwAABw8eVPqc6io++R5Ky2Xw59RVRERE9K+nhrbp06cjKysLISEh2Lt3LxITE5GWloaEhATs2rULgwcPxt27dxU6DTyLra0tZs2ahXHjxmHw4MEYOHAg2rdvj8mTJyM+Pv6Zr128eDH27t2L/v37Izo6Gu+991613qS6iRJLoKejCS9nS1WXQkRERA3EU59pMzY2xt69e7Fq1SosX74cRUVFACqeMTM1NcXAgQPxzjvvwMLCQumTBQcHIzg4WGHd119/XWk/e3t7nDx5Ur7cokULpXupqjtBEBAtlsDbxRraWpqqLoeIiIgaiGfOiGBqaoqPP/4YixYtQlpaGh48eABzc3O0atUKGhpKTVtK1fRPRj6ycosworerqkshIiKiBkSpaax0dHTg7Oxc17UQKppGAfB5NiIiIlLA22UNTLRYAqfmprA01Vd1KURERNSAMLQ1IAUPSyG+mcMJ4omIiKgShrYG5FJSJmQyAR3ZNEpERERPYGhrQKLEEhgb6MCllXIDFhMREVHTwdDWQEhlAmLEmfBzt4GmBmdBICIiIkUMbQ3E9X9ykf+wlL1GiYiIqEoMbQ1ElFgCDRHg696451UlIiKiF8PQ1kBEJ0rg3toCxgY6qi6FiIiIGiCGtgYg+34RUu/cZ9MoERERPRVDWwMQ/e8sCB097FRcCRERETVUDG0NQFSiBFZm+nCwM1Z1KURERNRAMbSpWFm5FLHXs9CxrS1EIg71QURERFVjaFOx+JRsFJdKOXUVERERPRNDm4pFiyXQ0dJA+zZWqi6FiIiIGjCGNhUSBAHRiRJ4tbGCno6WqsshIiKiBoyhTYVuZxXgbnYhJ4gnIiKi52JoU6FHQ334c6gPIiIieg6GNhWKSpSgpa0xbC0MVF0KERERNXAMbSrysLgMV1Kz2TRKRERESmFoU5G/r2VBKhM41AcREREphaFNRaITJTDU00Lb1haqLoWIiIjUAEObCshkAqKvSuDjZgMtTV4CIiIiej4mBhVIuZ2HvPwSdGTTKBERESmJoU0FohMlEIkAP3eGNiIiIlIOQ5sKRIklcG1pDlMjXVWXQkRERGqCoa2e5eYX43paHnuNEhERUbUwtNWzGHEmAMCf47MRERFRNTC01bNosQQWJrpwbmGq6lKIiIhIjTC01aNyqQx/X8uEn7stRCKRqsshIiIiNcLQVo8Sb2TjYXE5h/ogIiKiamNoq0dRiRJoaYrg7WKt6lKIiIhIzTC01aNosQSeTlYw0NNWdSlERESkZuo1tB0+fBj9+/fHq6++ip07d1ba/uuvvyI4OBgDBgzAvHnzUFpaCgA4cOAAunXrhkGDBmHQoEEIDw+vz7JrRUZ2IdIzCzjUBxEREb0Qrfo6kUQiQXh4OPbv3w8dHR2MHDkSnTt3Rps2bQAADx8+RFhYGA4cOAArKyvMmjULBw4cwIgRI5CQkIB58+Zh4MCB9VVurYtKlAAAOnKoDyIiInoB9Xan7dy5cwgICICZmRkMDAwQFBSEo0ePyrcbGBjg5MmTsLKyQlFREbKzs2FiYgIAiI+Px4EDBxAcHIwPPvgA9+/fr6+ya020WILmVoZobm2k6lKIiIhIDdXbnbbMzExYW//3AL6NjQ3i4uIU9tHW1sbp06cxZ84c2NjYoFu3bgAAa2trTJw4Eb6+vlizZg3CwsKwevVqpc+dkJBQO2/iBZWWyxB7PRMdXYwQExOj0lpqQp1rpwq8huqN10/98RqqP1Vew3oLbTKZTGFsMkEQqhyrrHv37rhw4QLWrFmDJUuWYPXq1diwYYN8+1tvvYU+ffpU69yenp7Q1VXdPJ8XEu5CKruDAa94oYOrjcrqqImYmBj4+fmpugyqAV5D9cbrp/54DdVfXV/DkpKSZ95oqrfmUTs7O2RlZcmXs7KyYGPzX4DJy8vD2bNn5cvBwcFISkpCfn4+tm7dKl8vCAI0NTXrpebaEiWWQF9XE+2crFRdChEREampegttXbt2xfnz55GTk4OioiIcP34cgYGB8u2CIGD27Nm4c+cOAODo0aPw9fWFgYEBNm/ejNjYWADAjh07qn2nTZUEQUCMWIIOrjbQ1uIIK0RERPRi6q151NbWFrNmzcK4ceNQVlaGYcOGoX379pg8eTJCQ0Ph5eWFZcuWYcqUKRCJRGjTpg2WLl0KTU1NrF27FkuWLEFxcTFat26NlStX1lfZNXbz7gPcu1+MN9hrlIiIiGqg3kIbUNHkGRwcrLDu66+/ln/du3dv9O7du9Lr/P39ceDAgTqvry48GurDn6GNiIiIaoDtdXUsWiyBs70pLEz0VF0KERERqTGGtjr0oLAUSbdyeJeNiIiIaoyhrQ5duiqBTOAsCERERFRzDG11KEosgamRDlxamqu6FCIiIlJzDG11RCqV4dLVTPi520JDo/IgwkRERETVwdBWR67eykVBURmfZyMiIqJawdBWR6LFEmhoiODjpp7TVhEREVHDwtBWR6LFEng4WsBIX1vVpRAREVEjwNBWBzJzH+Lm3QfsNUpERES1hqGtDsSIOQsCERER1S6GtjoQJZbAxsIALW2NVV0KERERNRIMbbWspEyK2Ov30LGtLUQiDvVBREREtYOhrZYlpNxDaZmUTaNERERUqxjaall0ogQ62prwamOl6lKIiIioEWFoq0WCICBKLIG3ixV0tTVVXQ4RERE1IgxttSg9swCSnIcc6oOIiIhqHUNbLYpKrBjqw4+hjYiIiGoZQ1stihZL0LqZCWzMDVRdChERETUyDG21pLCoDIk3stlrlIiIiOoEQ1st+ftaJqQygaGNiIiI6gRDWy2JSpTASF8b7g7mqi6FiIiIGiGGtlogkwm4dDUTvu420NTkR0pERES1jwmjFiSn5yGvoIRDfRAREVGdYWirBVGJEmiIAF93hjYiIiKqGwxttSBanAE3BwuYGOqouhQiIiJqpBjaauh+QQmS0++z1ygRERHVKYa2GjLU18ZrLzuhT+dWqi6FiIiIGjEtVReg7rQ0NTB5sJeqyyAiIqJGjnfaiIiIiNQAQxsRERGRGmBoIyIiIlIDDG1EREREaoChjYiIiEgNMLQRERERqYF6DW2HDx9G//798eqrr2Lnzp2Vtv/6668IDg7GgAEDMG/ePJSWlgIA7ty5g9GjR6Nv376YNm0aCgsL67NsIiIiIpWrt9AmkUgQHh6OXbt24eDBg9izZw+Sk5Pl2x8+fIiwsDBs2bIFP//8M0pKSnDgwAEAwNKlSzFq1CgcPXoUnp6eiIyMrK+yiYiIiBqEegtt586dQ0BAAMzMzGBgYICgoCAcPXpUvt3AwAAnT56ElZUVioqKkJ2dDRMTE5SVlSEqKgpBQUEAgJCQEIXXERERETUF9TYjQmZmJqytreXLNjY2iIuLU9hHW1sbp0+fxpw5c2BjY4Nu3bohNzcXRkZG0NKqKNXa2hoSiUSpcwqCAABISEiopXfRtMXExKi6BKohXkP1xuun/ngN1V99XMNH+eVJ9XanTSaTQSQSKRT0+PIj3bt3x4ULF9CjRw8sWbKkyv2qel1VysrKalY0ERERUT17Wn6ptzttdnZ2iI6Oli9nZWXBxsZGvpyXl4eEhAR069YNABAcHIxZs2bBwsIC+fn5kEql0NTUrPS6ZzE0NISrqyu0tbWVDnpEREREqiAIAsrKymBoaFjl9noLbV27dkVERARycnKgr6+P48ePY9myZQqFzp49G/v27UPz5s1x9OhR+Pr6QltbG/7+/jhy5AiCg4Nx8OBBBAYGKnVODQ0NGBsb19VbIiIiIqpVenp6T90mEp7WcFoHDh8+jE2bNqGsrAzDhg3D5MmTMXnyZISGhsLLywu//fYb1q1bB5FIhDZt2mDp0qUwNjbG7du3MW/ePGRnZ6NZs2ZYs2YNTE1N66tsIiIiIpWr19BGRERERC+GMyIQERERqQGGNiIiIiI1wNBGREREpAYY2oiIiIjUAEMbERERkRpgaKNnWr9+PQYMGIABAwZg5cqVqi6HamDFihWYN2+eqsugF3Dy5EmEhISgX79++Pjjj1VdDr2AQ4cOyf8tXbFiharLISUVFBRg4MCBSE9PB1Axj3pwcDBeffVVhIeH13s9DG30VOfOncPZs2dx4MABHDx4EFeuXMGvv/6q6rLoBZw/fx4HDhxQdRn0AtLS0rB48WJERkbif//7HxITE3H69GlVl0XVUFRUhE8++QTbt2/HoUOHEB0djXPnzqm6LHqO2NhYvPHGG7h58yYAoLi4GPPnz0dkZCSOHDmChISEev9ZZGijp7K2tsa8efOgo6MDbW1tODs7486dO6oui6opLy8P4eHhmDp1qqpLoRfw66+/on///rCzs4O2tjbCw8Ph7e2t6rKoGqRSKWQyGYqKilBeXo7y8nLo6uqquix6jr1792Lx4sXyqTPj4uLg4OCAli1bQktLC8HBwTh69Gi91lRv01iR+nFxcZF/ffPmTfzyyy/YvXu3CiuiF7Fo0SLMmjULd+/eVXUp9AJu3boFbW1tTJ06FXfv3sUrr7yC9957T9VlUTUYGRnh3XffRb9+/aCvr4+OHTvC19dX1WXRc3zyyScKy5mZmbC2tpYv29jYQCKR1GtNvNNGz3X9+nVMnDgRc+bMQevWrVVdDlXDDz/8gGbNmqFLly6qLoVekFQqxfnz5/Hpp59iz549iIuLY1O3mrl69Sr27duH33//HWfOnIGGhga++eYbVZdF1SSTySASieTLgiAoLNcHhjZ6ppiYGEyYMAHvv/8+hgwZoupyqJqOHDmCP//8E4MGDcIXX3yBkydP4tNPP1V1WVQNVlZW6NKlCywsLKCnp4fevXsjLi5O1WVRNZw9exZdunSBpaUldHR0EBISgosXL6q6LKomOzs7ZGVlyZezsrLkTaf1hc2j9FR3797FO++8g/DwcN6pUVNbtmyRf71//35cvHgR8+fPV2FFVF09evTA3Llz8eDBAxgaGuLMmTPo1auXqsuianB3d8eqVavw8OFD6Ovr4+TJk/Dy8lJ1WVRN3t7euHHjBm7dugV7e3v89NNPGDp0aL3WwNBGT/XNN9+gpKQEy5cvl68bOXIk3njjDRVWRdS0eHt746233sKoUaNQVlaGl156qd5/UVDNdOvWDYmJiQgJCYG2tja8vLzw9ttvq7osqiZdXV0sX74cM2fORElJCbp3746+ffvWaw0iQRCEej0jEREREVUbn2kjIiIiUgMMbURERERqgKGNiIiISA0wtBERERGpAYY2IiIiIjXA0EZEda5nz54YMWIEnuysfuHCBbi5uaG8vLzWzzl27FiEh4fX+nGVlZGRgeDgYHh5eWHPnj0K23r27Ak3N7en/peenv7C51X1+yaiusNx2oioXly+fBl79+7FiBEjVF1Kvfjmm28gEolw5MgRmJubK2z78ccfIZVKAQBff/014uLiEBERId9uYWHxwueNiIiAtrb2C7+eiBouhjYiqhctWrTAmjVr0KdPnxqFEnVRUFAAFxcXtGzZstK2x9+/vr4+tLW1FSairgkzM7NaOQ4RNTxsHiWiejFhwgQYGhpi1apVT93Hzc0N586dky/v378fgYGBACqaUgMDA7Fv3z689NJL6NixI7799ltcuHABffv2hY+PDz788EPIZDL56zMzMzF27Fh4eXlhxIgRuHnzpnxbfn4+5s6dCz8/P7z00ktYuHAhCgoKFM4VFhYGPz8/hbtgj8hkMmzevBm9e/dG+/btMWbMGFy9ehVARRPl/v378dNPP8HNza3an1VKSgomTZoEX19fdOvWDREREfL3FRERgdDQUMyfPx/e3t4ICgrCb7/9Jn/tk82j27dvR69eveDj44Nx48YhJSUFQMU0dW+99RZ8fX3RqVMnfPjhhygsLKx2rURUfxjaiKhe6OvrY/78+Thw4ABiYmJe6BjZ2dk4duwYtm3bhsmTJ+Pzzz/HihUrsGLFCqxcuRL/+9//cOrUKfn+Bw8eRFBQEA4ePAh7e3tMnDhR/vzc/PnzkZubi507d2LTpk24ceMGPvzwQ/lrJRIJCgoKcODAAQwZMqRSLRs2bMC3336LDz/8EAcOHIC9vT3eeustFBQUICIiAv369UNQUBDOnj1brfeYk5ODUaNGwcbGBj/88AOWLFmCnTt34ttvv5Xvc/LkSUilUuzfvx/Dhg1DaGgokpKSKh3rhx9+QHh4OGbNmoWDBw/Czs4O06dPhyAICAsLg5aWFvbt24dvv/0Wf//9N7788stq1UpE9YvNo0RUb3r37o1XXnkFS5cuxf79+6v9+vLycsyZMwfOzs6wtbXF6tWrMXr0aHh7ewMAnJ2dkZqaip49e8rPN2bMGADA0qVL8fLLL+PMmTNwdnbGr7/+ir/++kvenLhixQr07NkTd+/elZ/vrbfeQqtWrSrVIQgCduzYgXfffVc+efuyZcvQp08fHDp0CKNHj4aenh7Ky8ur3ez5008/QVdXF2FhYdDW1oazszOysrKwbt06vPXWWwAAExMTLFu2DDo6OnB2dsbp06fx448/4qOPPlI41vfff4+xY8di4MCBAICFCxciMjISBQUFuH37Ntzc3NCiRQvo6Ohg/fr1EIlE1aqViOoX77QRUb1asGABbt26he3bt7/Q6x89I6anpwcAaN68uXybnp4eSktL5cteXl7yr42MjODo6IiUlBSkpKRAEAT06NEDPj4+8PHxkQebx5tQW7RoUWUN2dnZyMvLk4dFANDW1oanp6e8+fFFpaSkwMPDQ6EzgY+PD3Jzc5GTkwMA8PDwgI6Ojny7p6cnUlNTqzyWp6enfNnY2Bhz586FsbExQkNDcfToUQQEBCA0NBRXr16Fo6NjjWonorrFO21EVK/s7e0xdepUREREYMmSJc/c91EPy8dpamoqLGtoPP1vzyfvHMlkMmhra0MqlcLAwAAHDx6s9Bpra2vEx8cDAHR1das87tPWS6XSKmuujqqO/eh5tkf/19JS/KdbKpVWeZfsWb1Ie/fujdOnT+O3337DH3/8gQ8//BBnz57F8uXLa1I+EdUh3mkjono3adIk2NjYYO3atQrrtbW15Z0BACAtLa1G57l27Zr86wcPHuDmzZtwdnaGo6MjHj58CKlUCgcHBzg4OAAAPvvsM4XzP42xsTGsra0RGxsrX1dWVoYrV67U+G6Vs7MzEhMTUVZWJl/3999/w8zMTN7r9Nq1awodLhISEqrs8ODg4IDExET58sOHD/HSSy/h2rVrCA8PR0ZGBl5//XWsX78eH3/8MY4cOVKj2omobjG0EVG909HRweLFi3H79m2F9V5eXtiyZQtu3ryJ33///YWee3vcL7/8gj179iA5ORnz589Hq1at0LVrVzg7O+Pll1/GnDlzEBsbi6tXr2Lu3LnIzs6GjY2NUseeOHEi1q9fjxMnTiAlJQWLFi1CSUmJvJn1RQ0cOBAymQyLFi1CSkoKTpw4gYiICIwcOVJ+V/H27dtYvnw5UlNT8eWXXyIhIQHDhw+vdKxx48Zh+/btOHbsGG7evInFixfDzMwMbdq0QWpqKsLCwpCYmIjU1FQcP34c7dq1q1HtRFS3GNqISCW6dOlSKeAsXLhQHnw2bdqEd999t0bneDT0xpAhQ/DgwQNs2LBBHnxWrlwJBwcHTJw4EWPGjIGNjQ0iIyOVPvaECRMwcuRILF68GCEhIbhz5w62b98OKyurGtVsaGiIzZs3Iy0tDYMHD0ZYWBjGjRun8Fl4enoiPz8fQ4YMwZEjR/DVV1+hdevWlY712muvYcqUKfjkk08wZMgQ5OTkYOPGjdDQ0MCSJUtga2uLCRMmICQkBFKpFKtXr65R7URUt0TCk/PKEBFRgxUREYFz585h9+7dqi6FiOoZ77QRERERqQGGNiIiIiI1wOZRIiIiIjXAO21EREREaoChjYiIiEgNMLQRERERqQGGNiIiIiI1wNBGREREpAYY2oiIiIjUwP8Dilk//LDpCzcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x_val = [x[0] for x in coherence]\n",
    "y_val = [x[1] for x in coherence]\n",
    "\n",
    "plt.figure(figsize=(10,5))\n",
    "ax = sns.lineplot(x=x_val, y=y_val, label='Topic Coherence')\n",
    "\n",
    "ax.axes.set_title('Coherence Score per Number of Topics', fontsize=16)\n",
    "ax.set_ylabel('Coherence Score', fontsize=14)\n",
    "ax.set_xlabel('Number of Topics', fontsize=14)\n",
    "plt.show()   "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1a3fe050-d072-4250-b361-d58bf2455ef8",
   "metadata": {},
   "source": [
    "The ideal **number of topics** will **maximize coherence**, in our case **8** can be a reasonable choice."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "380632d9-5069-4cda-9d10-04b940a15fe1",
   "metadata": {},
   "source": [
    "<a class=\"anchor\" id='chapter4.2.3'></a>\n",
    "### Build LDA Model and Interpret Results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "385b47da-19d6-41a1-9b86-93b236004295",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[(0, '0.057*\"view\" + 0.044*\"flat\" + 0.036*\"amazing\" + 0.031*\"lovely\" + 0.030*\"terrace\" + 0.022*\"enjoy\" + 0.021*\"wonderful\" + 0.020*\"love\" + 0.018*\"definitely\" + 0.017*\"fantastic\"'), (1, '0.024*\"distance\" + 0.019*\"attraction\" + 0.018*\"away\" + 0.016*\"area\" + 0.015*\"main\" + 0.015*\"tourist\" + 0.014*\"walking\" + 0.013*\"duomo\" + 0.013*\"shop\" + 0.012*\"right\"'), (2, '0.049*\"give\" + 0.022*\"provide\" + 0.021*\"tip\" + 0.019*\"information\" + 0.017*\"recommendation\" + 0.017*\"welcome\" + 0.016*\"eat\" + 0.016*\"visit\" + 0.015*\"lot\" + 0.015*\"map\"'), (3, '0.033*\"house\" + 0.029*\"bus\" + 0.025*\"quiet\" + 0.023*\"accommodation\" + 0.020*\"foot\" + 0.019*\"area\" + 0.016*\"parking\" + 0.016*\"pleasant\" + 0.016*\"car\" + 0.015*\"equip\"'), (4, '0.036*\"check\" + 0.029*\"easy\" + 0.019*\"communication\" + 0.018*\"super\" + 0.016*\"quick\" + 0.016*\"question\" + 0.014*\"definitely\" + 0.014*\"respond\" + 0.013*\"checkin\" + 0.012*\"accommodate\"'), (5, '0.015*\"air\" + 0.012*\"work\" + 0.012*\"kitchen\" + 0.011*\"floor\" + 0.010*\"bathroom\" + 0.009*\"stair\" + 0.009*\"bed\" + 0.009*\"water\" + 0.009*\"problem\" + 0.009*\"day\"'), (6, '0.031*\"night\" + 0.030*\"bed\" + 0.027*\"street\" + 0.024*\"bathroom\" + 0.021*\"small\" + 0.019*\"kitchen\" + 0.019*\"noise\" + 0.019*\"window\" + 0.018*\"bedroom\" + 0.017*\"little\"'), (7, '0.025*\"house\" + 0.023*\"welcome\" + 0.023*\"home\" + 0.020*\"feel\" + 0.017*\"kind\" + 0.016*\"like\" + 0.014*\"experience\" + 0.011*\"come\" + 0.010*\"return\" + 0.010*\"detail\"')]\n"
     ]
    }
   ],
   "source": [
    "# number of topics\n",
    "num_topics = 8\n",
    "\n",
    "# Build LDA model\n",
    "lda_model_final = gensim.models.LdaMulticore(corpus=corpus,\n",
    "                                       id2word=id2word,\n",
    "                                       num_topics=num_topics,\n",
    "                                       chunksize=200,\n",
    "                                       passes=40,\n",
    "                                       random_state=617,\n",
    "                                       per_word_topics=True)\n",
    "\n",
    "# Print the Keyword in the 8 topics\n",
    "print(lda_model_final.print_topics())\n",
    "doc_lda = lda_model_final[corpus]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "f0e57df6-3a34-4818-acbc-3d5482ffef10",
   "metadata": {},
   "outputs": [],
   "source": [
    "topics_topkeywords = dict(lda_model_final.print_topics())\n",
    "topics_topkeywords = {key: list(map(str, value.split('+'))) for key, value in ex.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "a82cd6df-56d2-48b4-a3b6-dc376688e6dd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# using dictionary comprehension to go through all keys\n",
    "topics_topkeywords_clean = {key: [re.sub(r'[\\d+\\W]', '', ele) for ele in val]\n",
    "       for key, val in ex2.items()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "08f441a8-266b-4191-9712-c8cf6f6aa2be",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "╒════════════╤════════════╤════════════════╤═══════════════╤═══════════════╤══════════╤══════════╤════════════╕\n",
      "│ 0          │ 1          │ 2              │ 3             │ 4             │ 5        │ 6        │ 7          │\n",
      "╞════════════╪════════════╪════════════════╪═══════════════╪═══════════════╪══════════╪══════════╪════════════╡\n",
      "│ view       │ distance   │ give           │ house         │ check         │ air      │ night    │ house      │\n",
      "├────────────┼────────────┼────────────────┼───────────────┼───────────────┼──────────┼──────────┼────────────┤\n",
      "│ flat       │ attraction │ provide        │ bus           │ easy          │ work     │ bed      │ welcome    │\n",
      "├────────────┼────────────┼────────────────┼───────────────┼───────────────┼──────────┼──────────┼────────────┤\n",
      "│ amazing    │ away       │ tip            │ quiet         │ communication │ kitchen  │ street   │ home       │\n",
      "├────────────┼────────────┼────────────────┼───────────────┼───────────────┼──────────┼──────────┼────────────┤\n",
      "│ lovely     │ area       │ information    │ accommodation │ super         │ floor    │ bathroom │ feel       │\n",
      "├────────────┼────────────┼────────────────┼───────────────┼───────────────┼──────────┼──────────┼────────────┤\n",
      "│ terrace    │ main       │ recommendation │ foot          │ quick         │ bathroom │ small    │ kind       │\n",
      "├────────────┼────────────┼────────────────┼───────────────┼───────────────┼──────────┼──────────┼────────────┤\n",
      "│ enjoy      │ tourist    │ welcome        │ area          │ question      │ stair    │ kitchen  │ like       │\n",
      "├────────────┼────────────┼────────────────┼───────────────┼───────────────┼──────────┼──────────┼────────────┤\n",
      "│ wonderful  │ walking    │ eat            │ parking       │ definitely    │ bed      │ noise    │ experience │\n",
      "├────────────┼────────────┼────────────────┼───────────────┼───────────────┼──────────┼──────────┼────────────┤\n",
      "│ love       │ duomo      │ visit          │ pleasant      │ respond       │ water    │ window   │ come       │\n",
      "├────────────┼────────────┼────────────────┼───────────────┼───────────────┼──────────┼──────────┼────────────┤\n",
      "│ definitely │ shop       │ lot            │ car           │ checkin       │ problem  │ bedroom  │ return     │\n",
      "├────────────┼────────────┼────────────────┼───────────────┼───────────────┼──────────┼──────────┼────────────┤\n",
      "│ fantastic  │ right      │ map            │ equip         │ accommodate   │ day      │ little   │ detail     │\n",
      "╘════════════╧════════════╧════════════════╧═══════════════╧═══════════════╧══════════╧══════════╧════════════╛\n"
     ]
    }
   ],
   "source": [
    "print(tabulate(topics_topkeywords_clean, headers='keys',tablefmt=\"fancy_grid\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2a1cfed6-e83e-41be-a366-970571450e46",
   "metadata": {},
   "source": [
    "#### **Topics** <br>\n",
    "\n",
    "Looking at the **top 10 keywords per topic**, we can summarize topics as follows: <br>\n",
    "\n",
    "1. **Good feelings**\n",
    "2. **Location**\n",
    "3. **Tips**\n",
    "4. **Accommodation benefits**\n",
    "5. **Host communication**\n",
    "6. **Accommodation features**\n",
    "7. **Complaints**\n",
    "8. **Good experience**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "0c7f6d66-46d8-4cc6-81e6-8c5cd7ccfbdb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a dictionary with topics\n",
    "\n",
    "topic_list = {0: 'Good feelings', \n",
    "              1: 'Location', \n",
    "              2: 'Tips', \n",
    "              3: 'Accommodation benefits', \n",
    "              4: 'Host communication', \n",
    "              5: 'Accommodation features', \n",
    "              6: 'Complaints', \n",
    "              7: 'Good experience'\n",
    "             }"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "b3247c75-cf20-4fbb-b9e4-3cd5a6ca6fc5",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Add main and secondary topic to dataframe\n",
    "\n",
    "N = len(df_reviews_sample_sentiment_bitrigrams)\n",
    "main_topic = -np.ones(N, int)\n",
    "secondary_topic = -np.ones(N, int)\n",
    "\n",
    "for i in range(N):\n",
    "    results = sorted(lda_model_final.get_document_topics(corpus[i]), key=lambda x: x[1], reverse=True)\n",
    "    main_topic[i] = results[0][0]\n",
    "    if len(results) > 1:\n",
    "        secondary_topic[i] = results[1][0]\n",
    "        \n",
    "df_reviews_sample_sentiment_bitrigrams['main_topic'] = main_topic\n",
    "df_reviews_sample_sentiment_bitrigrams['secondary_topic'] = secondary_topic"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "5ef8a860-26e9-4990-9638-e57b7c462ab1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['marzia is a great host the accommodation is perfect and is lovingly furnished to feel good with an eye for practical and beautiful details the communication went smoothly and a prosecco was already waiting for us in the fridge we can highly recommend the accommodation and would be happy to come back',\n",
       "       7, 4], dtype=object)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check some examples\n",
    "\n",
    "df_reviews_sample_sentiment_bitrigrams[['comments_english','main_topic', 'secondary_topic']].iloc[13].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "83bdb6da-4d95-443f-9b78-e60b45753503",
   "metadata": {},
   "outputs": [],
   "source": [
    "main_topic_dis = pd.DataFrame(df_reviews_sample_sentiment_bitrigrams['main_topic'].value_counts(normalize=True).rename(index=topic_list).reset_index())\n",
    "main_topic_dis = main_topic_dis.rename(columns={'index': 'Topic','main_topic': 'percentage'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "de8a78ed-d1c8-435d-a9e7-93e9081bb2ca",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Topic</th>\n",
       "      <th>percentage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Good experience</td>\n",
       "      <td>16.114561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Location</td>\n",
       "      <td>15.358395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Host communication</td>\n",
       "      <td>13.765745</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Accommodation benefits</td>\n",
       "      <td>13.548299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Good feelings</td>\n",
       "      <td>12.280839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Accommodation features</td>\n",
       "      <td>12.259290</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Tips</td>\n",
       "      <td>9.338453</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Complaints</td>\n",
       "      <td>7.334417</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    Topic  percentage\n",
       "0         Good experience   16.114561\n",
       "1                Location   15.358395\n",
       "2      Host communication   13.765745\n",
       "3  Accommodation benefits   13.548299\n",
       "4           Good feelings   12.280839\n",
       "5  Accommodation features   12.259290\n",
       "6                    Tips    9.338453\n",
       "7              Complaints    7.334417"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "main_topic_dis['percentage'] = main_topic_dis['percentage']*100\n",
    "main_topic_dis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "17cae0bc-730a-45cb-a407-9bf6e5405277",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(6,5))\n",
    "\n",
    "x = main_topic_dis['Topic']\n",
    "y = main_topic_dis['percentage']\n",
    "\n",
    "ax = sns.barplot(x=\"percentage\", y=\"Topic\", data=main_topic_dis,\n",
    "            label=\"Total\", palette=\"Blues_r\")\n",
    "\n",
    "ax.set_title('Main Topics Distribution (total sum 100%)', fontsize=16)\n",
    "ax.set_ylabel('Topic', fontsize=15)\n",
    "ax.set_xlabel('% of reviews', fontsize=15)\n",
    "\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "fa2d8c08-dad5-4245-8524-112a36858e18",
   "metadata": {},
   "outputs": [],
   "source": [
    "ax.figure.savefig('main_topic_distribution.png',dpi=600, bbox_inches = \"tight\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dc4e5f49-686d-40a2-aecd-07fe73751647",
   "metadata": {},
   "source": [
    "Looking at the **main topic distribution** across our sample of reviews: <br>\n",
    "- The most discussed topics are **Good experience** and **Location**\n",
    "- The least discussed topics are **Tips** and **Complaints**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "716fe917-f4aa-4917-ba16-12f7c4c2a9c8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Remove some outliers in lenght_comments\n",
    "\n",
    "quantile_95 = df_reviews_sample_sentiment_bitrigrams[\"length_comments\"].quantile(0.95)\n",
    "df_95 = df_reviews_sample_sentiment_bitrigrams[df_reviews_sample_sentiment_bitrigrams[\"length_comments\"] < quantile_95]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "65fc67bc-3d2d-464e-bbcc-1ec75ec8ce68",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y = df_95[\"main_topic\"].map(topic_list)\n",
    "x = df_95[\"length_comments\"]\n",
    "\n",
    "plt.figure(figsize=(10,8))\n",
    "\n",
    "ax = sns.boxplot(x=x, y=y, data=df_reviews_sample_sentiment_bitrigrams, palette=\"Set2\" )\n",
    "\n",
    "ax.set_title('Length Comments Distribution by Topic', fontsize=16)\n",
    "ax.set_ylabel('Topic', fontsize=15)\n",
    "ax.set_xlabel('Number of Words', fontsize=15)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "3861108e-38f8-41cb-84b2-decbe116ab76",
   "metadata": {},
   "outputs": [],
   "source": [
    "ax.figure.savefig('length_comments_topic.png',dpi=600, bbox_inches = \"tight\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "65f06707-6732-456e-8770-e96e53be8fcb",
   "metadata": {},
   "source": [
    "Looking at the **length of comments distribution** by topics: <br>\n",
    "- The most **wordy** topic is **Accommodation features**\n",
    "- The most **concise** topic is **Good feelings**\n",
    "- The most **diverse** topics are **Accommodation features**, **Location**, **Complaints**\n",
    "- Most topics are **right-skewed**\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1a6fde60-33e3-418d-a9d8-65854553f27c",
   "metadata": {},
   "source": [
    "<a class=\"anchor\" id='chapter4_2_4'></a>\n",
    "### Evaluate LDA Model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "015748a7-52e9-4a35-a111-17057bcbcf6c",
   "metadata": {},
   "source": [
    "Like each algorithm LDA also needs to evaluate how good the trained model is. There are two convenient measurement techniques:<br>\n",
    "* **Perplexity Score**: Lower is better\n",
    "* **Coherence Score**: Higher is better"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "430cfd72-7c52-4a73-822d-fb541d840eae",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Perplexity:  -7.1401473452000115\n",
      "\n",
      "Coherence Score:  0.5993569972468963\n"
     ]
    }
   ],
   "source": [
    "# Compute Perplexity Score\n",
    "print('\\nPerplexity: ', lda_model_final.log_perplexity(corpus))\n",
    "\n",
    "# Compute Coherence Score\n",
    "coherence_model_lda = CoherenceModel(model=lda_model_final, texts=data_words, dictionary=id2word, coherence='c_v')\n",
    "coherence_lda = coherence_model_lda.get_coherence()\n",
    "print('\\nCoherence Score: ', coherence_lda)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "865d77f3-f834-45ba-8959-3baea1175012",
   "metadata": {},
   "source": [
    "[Back to TOC](#Table-of-Contents:)\n",
    "<a class=\"anchor\" id='chapter4.3'></a>\n",
    "### Visualize topic-keyword of LDA"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2ac872e-4b42-499a-9a0e-2d57fc9c56d1",
   "metadata": {},
   "source": [
    "The next step os to examine the topics from LDA and the associated keywords. <br>\n",
    "We use the **pyLDAvis** library for interactive topic visualization. pyLDAvis is **designed to help users interpret the topics in a topic model** that has been fit to a corpus of text data. The package extracts information from a fitted LDA topic model to inform an **interactive web-based visualization**. <br>\n",
    "\n",
    "pyLDAvis attempts to aswer the following questions: <br>\n",
    "1. What is the meaning of each topic?\n",
    "2. How prevalent is each topic?\n",
    "3. How do the topics relate to each other?\n",
    "\n",
    "**Reference:**<br>\n",
    "Sievert, Carson, and Kenneth Shirley. \"LDAvis: A method for visualizing and interpreting topics.\" Proceedings of the workshop on interactive language learning, visualization, and interfaces. 2014. https://nlp.stanford.edu/events/illvi2014/papers/sievert-illvi2014.pdf <br>\n",
    "https://pypi.org/project/pyLDAvis/\n",
    "\n",
    "**Packages** required: <br>\n",
    "- pyLDAvis\n",
    "- reload\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "7d27a150-408e-475b-bfd1-22048f3a0674",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/loredanafattorini/miniforge3/lib/python3.9/site-packages/pyLDAvis/_prepare.py:246: FutureWarning: In a future version of pandas all arguments of DataFrame.drop except for the argument 'labels' will be keyword-only\n",
      "  default_term_info = default_term_info.sort_values(\n",
      "/Users/loredanafattorini/miniforge3/lib/python3.9/site-packages/past/builtins/misc.py:45: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses\n",
      "  from imp import reload\n",
      "/Users/loredanafattorini/miniforge3/lib/python3.9/site-packages/past/builtins/misc.py:45: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses\n",
      "  from imp import reload\n",
      "/Users/loredanafattorini/miniforge3/lib/python3.9/site-packages/past/builtins/misc.py:45: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses\n",
      "  from imp import reload\n",
      "/Users/loredanafattorini/miniforge3/lib/python3.9/site-packages/past/builtins/misc.py:45: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses\n",
      "  from imp import reload\n",
      "/Users/loredanafattorini/miniforge3/lib/python3.9/site-packages/past/builtins/misc.py:45: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses\n",
      "  from imp import reload\n",
      "/Users/loredanafattorini/miniforge3/lib/python3.9/site-packages/past/builtins/misc.py:45: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses\n",
      "  from imp import reload\n",
      "/Users/loredanafattorini/miniforge3/lib/python3.9/site-packages/past/builtins/misc.py:45: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses\n",
      "  from imp import reload\n",
      "/Users/loredanafattorini/miniforge3/lib/python3.9/site-packages/past/builtins/misc.py:45: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses\n",
      "  from imp import reload\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "<link rel=\"stylesheet\" type=\"text/css\" href=\"https://cdn.jsdelivr.net/gh/bmabey/pyLDAvis@3.3.1/pyLDAvis/js/ldavis.v1.0.0.css\">\n",
       "\n",
       "\n",
       "<div id=\"ldavis_el7713859980289449921707526\"></div>\n",
       "<script type=\"text/javascript\">\n",
       "\n",
       "var ldavis_el7713859980289449921707526_data = {\"mdsDat\": {\"x\": [-0.11658206737915255, -0.033067779078088125, -0.16542654938890736, 0.04879868803731442, -0.13112889570016356, 0.19882753793843097, 0.31200809335549107, -0.11342902778492443], \"y\": [0.2879551028456152, 0.13550643240454913, -0.15847157764949651, -0.12620205459597275, -0.0004258660784363789, -0.10334071324207522, 0.08112382822177108, -0.11614515190595452], \"topics\": [1, 2, 3, 4, 5, 6, 7, 8], \"cluster\": [1, 1, 1, 1, 1, 1, 1, 1], \"Freq\": [8.874749379914212, 15.833814462651258, 11.153423474539888, 10.432188197718036, 12.780142404451164, 17.01253838770237, 10.395133309247953, 13.518010383775128]}, \"tinfo\": {\"Term\": [\"view\", \"give\", \"house\", \"flat\", \"check\", \"amazing\", \"bus\", \"street\", \"easy\", \"welcome\", \"bed\", \"distance\", \"terrace\", \"lovely\", \"home\", \"bathroom\", \"night\", \"accommodation\", \"provide\", \"quiet\", \"tip\", \"attraction\", \"area\", \"communication\", \"feel\", \"information\", \"kitchen\", \"love\", \"kind\", \"foot\", \"view\", \"flat\", \"terrace\", \"chiara\", \"luca\", \"spectacular\", \"rooftop\", \"roof\", \"guido\", \"stunning\", \"dome\", \"mercato\", \"thoroughly\", \"centrale\", \"sara\", \"lorenzos\", \"martina\", \"sunset\", \"marina\", \"giuseppe\", \"breathtaking\", \"lucas\", \"alice\", \"stroll\", \"francescos\", \"breathtake\", \"gian\", \"mirko\", \"tuck\", \"deck\", \"balcony\", \"gorgeous\", \"amazing\", \"lovely\", \"worth\", \"brilliant\", \"enjoy\", \"centre\", \"fantastic\", \"love\", \"absolutely\", \"incredible\", \"wonderful\", \"heart\", \"duomo\", \"definitely\", \"stylish\", \"space\", \"decorate\", \"explore\", \"spacious\", \"come\", \"firenze\", \"right\", \"central\", \"want\", \"vecchio\", \"ponte\", \"piazza\", \"river\", \"cafe\", \"uffizi\", \"bridge\", \"michele\", \"pitti\", \"gelato\", \"palazzo\", \"palace\", \"arno\", \"crowd\", \"touristy\", \"gallery\", \"spirito\", \"santo\", \"della\", \"signoria\", \"barbara\", \"bakery\", \"pizza\", \"bustle\", \"famous\", \"michelangelo\", \"claudio\", \"hustle\", \"irene\", \"boboli\", \"major\", \"walking\", \"distance\", \"attraction\", \"tourist\", \"shop\", \"grocery\", \"away\", \"main\", \"market\", \"site\", \"museum\", \"corner\", \"block\", \"sight\", \"bar\", \"duomo\", \"right\", \"neighborhood\", \"area\", \"local\", \"min\", \"nearby\", \"train\", \"quiet\", \"street\", \"food\", \"near\", \"easy\", \"feel\", \"love\", \"want\", \"definitely\", \"tip\", \"information\", \"map\", \"precise\", \"giulia\", \"give\", \"roberto\", \"tommaso\", \"father\", \"splendid\", \"organized\", \"ice\", \"antonio\", \"bottle\", \"cream\", \"info\", \"luigi\", \"sandra\", \"valuable\", \"elisa\", \"chianti\", \"biscuit\", \"giorgio\", \"alessandros\", \"cristiana\", \"tiziana\", \"nest\", \"breditz\", \"complimentary\", \"boiler\", \"camilla\", \"detailed\", \"massimo\", \"recommendation\", \"provide\", \"wine\", \"daniele\", \"useful\", \"eat\", \"guide\", \"explain\", \"greet\", \"snack\", \"suggestion\", \"advice\", \"arrival\", \"tour\", \"local\", \"direction\", \"marco\", \"show\", \"meet\", \"lot\", \"welcome\", \"visit\", \"help\", \"leave\", \"arrive\", \"breakfast\", \"receive\", \"thing\", \"wait\", \"friendly\", \"wonderful\", \"train\", \"kind\", \"bus\", \"parking\", \"car\", \"stop\", \"maria\", \"position\", \"transport\", \"tram\", \"public\", \"park\", \"novella\", \"strategic\", \"garage\", \"repeat\", \"reachable\", \"fabio\", \"fiore\", \"bril\", \"brje\", \"connected\", \"structure\", \"isabella\", \"conform\", \"internal\", \"countryside\", \"importantly\", \"giancarlo\", \"brlappartement\", \"deserve\", \"independent\", \"foot\", \"reach\", \"meter\", \"monument\", \"accommodation\", \"cathedral\", \"appartment\", \"accessible\", \"money\", \"pleasant\", \"house\", \"ideally\", \"value\", \"quiet\", \"min\", \"equip\", \"historic\", \"convenient\", \"near\", \"area\", \"supermarket\", \"kind\", \"available\", \"friendly\", \"central\", \"spacious\", \"visit\", \"welcome\", \"easy\", \"quick\", \"question\", \"respond\", \"checkin\", \"answer\", \"leonardo\", \"response\", \"silvia\", \"reply\", \"prompt\", \"process\", \"riccardo\", \"smooth\", \"max\", \"ilaria\", \"andreas\", \"annalisa\", \"query\", \"smoothly\", \"nicoletta\", \"samuel\", \"azadeh\", \"solo\", \"ample\", \"timely\", \"valeria\", \"giulio\", \"hopefully\", \"bite\", \"extreme\", \"check\", \"communicate\", \"flexible\", \"checkout\", \"accommodate\", \"federico\", \"communication\", \"drop\", \"message\", \"self\", \"easy\", \"early\", \"fast\", \"quickly\", \"let\", \"exactly\", \"super\", \"responsive\", \"describe\", \"meet\", \"definitely\", \"extremely\", \"late\", \"train\", \"arrive\", \"help\", \"amazing\", \"airbnb\", \"book\", \"go\", \"experience\", \"friendly\", \"love\", \"spacious\", \"wonderful\", \"air\", \"conditioning\", \"machine\", \"washing\", \"towel\", \"heat\", \"toilet\", \"lift\", \"internet\", \"fix\", \"conditioner\", \"dryer\", \"steep\", \"clothe\", \"soap\", \"dishwasher\", \"sheet\", \"dish\", \"cost\", \"microwave\", \"wash\", \"hair\", \"correct\", \"strong\", \"shampoo\", \"paper\", \"lock\", \"pressure\", \"system\", \"iron\", \"elevator\", \"hot\", \"water\", \"fridge\", \"wifi\", \"stair\", \"work\", \"floor\", \"suitcase\", \"issue\", \"shower\", \"problem\", \"summer\", \"use\", \"say\", \"kitchen\", \"difficult\", \"cold\", \"luggage\", \"bathroom\", \"coffee\", \"door\", \"leave\", \"bed\", \"thing\", \"day\", \"find\", \"small\", \"building\", \"like\", \"bit\", \"provide\", \"noise\", \"sleep\", \"noisy\", \"hear\", \"ceiling\", \"mosquito\", \"loud\", \"double\", \"earplug\", \"sleeper\", \"possibility\", \"face\", \"antonella\", \"oil\", \"winter\", \"mezzanine\", \"ztl\", \"natural\", \"ear\", \"brown\", \"tall\", \"olive\", \"midnight\", \"annoying\", \"disturb\", \"beam\", \"carmine\", \"shutter\", \"green\", \"edoardo\", \"window\", \"sound\", \"living\", \"bedroom\", \"sofa\", \"bit\", \"street\", \"night\", \"bed\", \"open\", \"light\", \"bother\", \"bathroom\", \"small\", \"kitchen\", \"little\", \"people\", \"large\", \"courtyard\", \"big\", \"fine\", \"outside\", \"morning\", \"space\", \"busy\", \"area\", \"quiet\", \"lot\", \"shower\", \"problem\", \"central\", \"comfort\", \"attention\", \"impeccable\", \"availability\", \"kindness\", \"magnificent\", \"hesitation\", \"maurizio\", \"mario\", \"casa\", \"brla\", \"elisabetta\", \"cat\", \"dog\", \"federica\", \"caring\", \"maximum\", \"surely\", \"exceed\", \"beauty\", \"carla\", \"patrizia\", \"reality\", \"discreet\", \"appreciable\", \"giovanna\", \"jacopo\", \"characteristic\", \"roberta\", \"delighted\", \"expectation\", \"hospitality\", \"detail\", \"doubt\", \"home\", \"soon\", \"taste\", \"pleasure\", \"care\", \"feel\", \"moment\", \"special\", \"cleanliness\", \"welcome\", \"house\", \"experience\", \"return\", \"warm\", \"like\", \"kind\", \"hope\", \"attentive\", \"decoration\", \"hostess\", \"come\", \"airbnb\", \"family\", \"spend\", \"guest\", \"wonderful\", \"day\", \"accommodation\", \"visit\", \"friendly\", \"super\", \"available\", \"photo\", \"help\", \"equip\"], \"Freq\": [4253.0, 4642.0, 5799.0, 3305.0, 4089.0, 4536.0, 2571.0, 3621.0, 5033.0, 5205.0, 4534.0, 3425.0, 2224.0, 4202.0, 3398.0, 3854.0, 4448.0, 2834.0, 2777.0, 5100.0, 1970.0, 2756.0, 5093.0, 2578.0, 3608.0, 1803.0, 3933.0, 3929.0, 4441.0, 1886.0, 4253.021420114006, 3304.2383114591685, 2223.4019095053595, 792.7973265443779, 620.207599360085, 448.683127280371, 366.33833400064884, 348.6352367975311, 345.00276306643065, 343.6328894983461, 292.6572564286186, 231.99134202992727, 224.08453743892147, 221.9765203018852, 203.89126825212443, 190.10976646142018, 179.324250925532, 176.80007697279814, 166.32010690883115, 159.80825758530506, 137.52252051583883, 128.88444621073643, 116.53429788586713, 106.51024980048128, 103.49571463199113, 102.18941110327337, 94.57920547844215, 92.59513636285928, 92.5670200227074, 85.30796693477157, 908.1029010114768, 394.80665749886094, 2687.5539737443933, 2309.7060431816617, 545.830705773173, 204.40937850281136, 1688.5528952376817, 701.9938088518815, 1250.6335600129696, 1533.3280296198925, 917.4681347103425, 472.183770095996, 1594.5763501737176, 728.2786602261738, 1114.8854560396308, 1353.2478645489894, 536.8920708527238, 814.8691520492671, 462.815213159414, 427.0730219293416, 717.0294655752457, 639.1108751190611, 440.6009781730214, 475.3754075986746, 481.4404563566144, 442.6714951960386, 1625.8105961970844, 1459.0772617937255, 1141.382199601418, 813.5849634594452, 781.125481436922, 630.9636242875818, 572.3797155954376, 473.8615350018685, 451.6435614485896, 433.06610399578426, 431.59168576934036, 391.69545390808725, 389.9299466785899, 374.450583686992, 363.7714743290449, 355.6008856185522, 351.5174948256114, 333.99202079711296, 330.966417733214, 320.56004828593234, 276.92203705073786, 267.1382260951925, 265.2996345555374, 228.6467295976241, 217.56105365774292, 207.64578768232593, 195.4117175682522, 178.93621418096703, 174.15978628207506, 156.142100970862, 1170.2749051583537, 1817.9846351370347, 3252.2185988987203, 2530.7797192298317, 1986.207116970251, 1689.6462148486914, 567.8498614269944, 2391.846882503169, 2034.6257533031917, 1279.7686827404455, 1233.344230761185, 618.2580012226997, 711.5869149416161, 669.3921159934301, 1002.7455072399976, 964.3007260337639, 1798.0009756413913, 1651.8693031582545, 1099.0896636509203, 2100.348707171852, 1174.4111900186374, 1183.6349167731962, 941.5129008204551, 1540.536621977113, 1633.5197114752063, 1278.6637428369168, 854.9743371108035, 1032.7376604551212, 1222.435133629599, 1048.5538915629243, 1031.900654747585, 945.3462384207, 956.5975398841347, 1969.5231342913203, 1802.3654964686173, 1404.341476901557, 483.8236425983856, 326.57358721415346, 4628.9054475487965, 282.975814969822, 267.4363629681496, 246.0137913677288, 230.09603550675968, 220.31016535584985, 206.7530387658706, 200.74024952935397, 1220.3112671379154, 185.39324670189092, 179.72301919362997, 123.98055002109216, 118.53378776754421, 117.30406941771055, 116.9421748455123, 170.61569999173997, 113.6331261490014, 113.55203591443163, 99.99544787439733, 99.41227521971155, 92.84905496699969, 81.29293817455913, 75.61240128163752, 74.17156585797174, 74.31787558995774, 146.92458017010233, 222.08990806171622, 240.05628593168467, 1619.0770348414144, 2120.748310542312, 1381.5546154099536, 409.8613913205527, 675.6082681572852, 1490.1725336643638, 374.7757196899626, 679.5286919402935, 713.0934985263151, 298.6002699362765, 640.2254488468594, 869.0997673172534, 1240.2830731131344, 436.52409784869985, 1185.630497139414, 400.2485605642032, 416.4921061694278, 707.1215731407242, 1217.1941840197449, 1427.493936870082, 1617.9865701202061, 1467.4661249400483, 1037.7259539041572, 989.574313233392, 930.255037124361, 715.5505172466377, 506.75184597350307, 689.210193310033, 544.7677568638079, 724.64119635431, 704.189531387316, 671.8739408572987, 630.497126604354, 2570.3025406485367, 1454.995956640869, 1381.8635254496726, 1201.8419563087589, 843.4481564651844, 840.642266221539, 753.40834090651, 722.3624540732794, 700.8937504437314, 643.9668198553022, 458.18095473800565, 359.80731290303453, 345.90866561330586, 235.4658221908716, 217.97792085999222, 201.86321405606958, 121.82762712655442, 121.43812460793711, 110.8802820239335, 108.37963864046655, 98.38102294316525, 94.8733414556641, 91.91925336281736, 91.43280954855854, 72.91915520073772, 69.50600432660578, 65.60820865483613, 65.62726795092514, 65.12043870832517, 64.01498925506762, 1752.5076621542207, 1130.432469035857, 316.1107823103912, 237.2931090731917, 2060.189917953379, 465.7493251422486, 490.5891597577064, 382.9714097963363, 417.18714335580967, 1400.2522303830847, 2938.1751637470456, 410.6947699327126, 569.2258005223453, 2201.6461631089137, 1225.4180794087745, 1301.844417512407, 758.0579045038704, 1051.5528068259407, 1093.3195991097143, 1654.17106978155, 747.4926808116726, 1301.3834580535984, 932.9292283016205, 1118.4945648214514, 1092.11271595931, 1039.6311040412204, 994.3063297652188, 935.2720208267019, 707.7926810630141, 1737.6744313159654, 1698.4764251909594, 1463.257256523244, 1413.6691491859483, 1239.0684587663852, 470.96274390994546, 445.0657236755918, 427.37443213824434, 336.0491027494313, 284.1701547609022, 279.8186380731288, 247.78041888656554, 232.8484076485217, 229.32143340182947, 224.21443876996423, 162.97222650605812, 145.72960232278504, 140.63360209388694, 132.9760570994617, 120.79866644812374, 109.11256063040284, 105.18165856748355, 104.96062447071827, 98.56307835654313, 98.19520664076842, 86.47318955150024, 84.74240857306322, 79.82652512679499, 74.05314751508877, 73.87641217254672, 3887.9275738287065, 857.7438756579672, 532.1259803422286, 445.042802927138, 1259.8099729825858, 340.77808134525833, 2087.4313506472863, 293.1249133672704, 621.1166677813893, 170.45528415442425, 3101.0944488542646, 937.3651152571955, 560.0917618441703, 779.5172754454284, 685.1156289872149, 847.8139021739787, 1898.7805940524377, 830.8542610773853, 598.5414433136128, 1217.695440148829, 1528.1238814024068, 832.8082674768904, 628.1820514613651, 1247.8524955782261, 958.4907912925361, 998.4902679843555, 1183.0878302482292, 856.858177637495, 729.0749546322397, 769.1215825780212, 880.5588360419891, 787.7481917780981, 803.1368800383573, 822.2652773161229, 719.0964697619632, 2097.7066358515312, 1218.473734077501, 1027.1371767986407, 810.1090098430262, 805.9817219020542, 498.7534231877311, 441.9879918254119, 390.40041311010276, 341.8270930700827, 318.4139813080998, 287.4578129450219, 281.44713303176593, 268.55535991036527, 267.8354080816095, 265.8067181118909, 246.575261592793, 242.56579461939577, 240.72915886204711, 230.90633024351493, 222.26893680400588, 221.1671732882106, 209.52713897798415, 199.97676715584126, 195.3462589553802, 194.7523762325097, 188.34040427303356, 188.06231460170045, 185.13563372018066, 184.1979947451589, 178.52114649486754, 783.3083943826919, 1234.2903104554402, 1344.0191167090766, 381.9165993353564, 1115.6445961077168, 1363.1954347588585, 1770.7603578093474, 1640.1393525051883, 422.5272471199851, 766.1227611612666, 1307.9901756399047, 1331.0905957844711, 462.55505999301704, 855.8577448598832, 388.93432328990605, 1663.945625526642, 402.55002142863896, 449.37953079632194, 807.3921535531633, 1500.2592419541058, 779.8858193320116, 781.9166719233604, 1064.5366499880697, 1346.4522504526626, 938.2714664083172, 1322.6614007792691, 910.3222809237259, 995.3431518166885, 655.1097264784763, 741.6640134588214, 626.7650327031874, 639.6586834495695, 1667.0792689901864, 1474.2347731484083, 1234.1822866798586, 597.2126972149628, 542.369346688539, 410.88434768696675, 348.89704229604877, 302.1566506237693, 251.93256069444897, 230.71740928623706, 216.55803600311637, 209.62772124979654, 193.28021677431124, 185.6063000383308, 163.0284348964386, 131.3588083355981, 129.10454458868128, 125.48509840522676, 120.18191492441272, 119.51071456746081, 113.78297565666342, 104.96015943904251, 101.40060746972476, 99.60774531120583, 89.04054980072519, 87.30563557608205, 87.20675429524363, 85.1233585898709, 83.29696919296141, 77.82753240353787, 1642.612190573276, 203.451211085137, 728.2915177696476, 1588.2666370761306, 418.5571171419512, 1434.50197379939, 2342.058277850899, 2698.189705338515, 2630.0998894071763, 639.9904213015529, 701.8057164917211, 240.06664309655318, 2090.0142465153967, 1852.272220075405, 1682.7798094197592, 1482.2752461625946, 1082.3952080455654, 1044.7245643381245, 325.3327101256658, 706.15612407972, 378.97981534220554, 570.9166889932234, 526.2022726368275, 698.6409418643935, 361.451697260406, 784.4807518272843, 774.3900876733356, 632.480647696636, 502.3779174235767, 504.650066461734, 494.6877338820466, 971.2862132758981, 826.0723231785065, 679.0496641572084, 532.6221842109129, 502.52764963586986, 335.76713142355857, 310.331085314882, 190.6080969087734, 190.16540249965175, 189.6093203491151, 172.5726789922333, 168.76002950808325, 166.78514860867833, 163.48202463076916, 154.83682211057808, 147.89572598089367, 143.1550207452587, 137.99677101307154, 128.29719037863262, 123.09091870004028, 121.43498078614975, 118.62293282883067, 111.9177769806244, 101.71546451336113, 101.50290246099286, 97.54067893378853, 96.01415383801904, 94.14320282095899, 93.79198322625186, 93.19014368711554, 344.92830677927213, 977.30961733629, 1136.7733839550965, 421.5049150936099, 2604.68351802603, 542.5586908755218, 370.87816773309186, 443.8860993645031, 660.3338665084655, 2294.933358392904, 242.70941774481673, 433.3709582178003, 590.4427390312826, 2651.672293234773, 2860.68597001479, 1632.4519451128272, 1170.2927824714877, 985.9543669449199, 1819.0846507568037, 1954.6526968443884, 527.8436475389976, 715.1085644385607, 290.6533035103562, 433.51389177811313, 1205.7927366027257, 922.7781062726505, 828.9797489822225, 701.3373355116622, 601.7080349324547, 1115.8226542453058, 952.6785815646892, 773.459825041117, 905.6351372663356, 799.120781702666, 746.734632778402, 670.0746088112518, 631.5843194187592, 641.2135389100879, 629.9688105221612], \"Total\": [4253.0, 4642.0, 5799.0, 3305.0, 4089.0, 4536.0, 2571.0, 3621.0, 5033.0, 5205.0, 4534.0, 3425.0, 2224.0, 4202.0, 3398.0, 3854.0, 4448.0, 2834.0, 2777.0, 5100.0, 1970.0, 2756.0, 5093.0, 2578.0, 3608.0, 1803.0, 3933.0, 3929.0, 4441.0, 1886.0, 4253.929803671106, 3305.1466670469854, 2224.3105602964524, 793.7058428755008, 621.116118768227, 449.59177086450103, 367.24643985985284, 349.5436070056034, 345.911319623729, 344.54106107620237, 293.5660780120508, 232.89972296465493, 224.9929305737911, 222.88482746483757, 204.80027266793005, 191.01796678369809, 180.23289903185952, 177.7083791757457, 167.22873260863452, 160.71690175660146, 138.4308391596833, 129.7926064400983, 117.4431175111244, 107.41915073037771, 104.40406667909005, 103.09744680379335, 95.48769871057351, 93.50407791491325, 93.47578247081982, 86.2161363931357, 1070.683919295184, 486.1837512293986, 4536.1907143594035, 4202.622949399755, 758.5491577951252, 238.79620267011285, 3786.940988234765, 1315.171355935854, 2929.4203748150935, 3929.4303920301236, 1983.0864783749587, 808.442278966477, 5027.04659121248, 1675.7432993375216, 3506.761659392054, 4855.630177381934, 1116.1044448637704, 3258.029387749762, 1308.284005058909, 1104.7583144952844, 4592.672118407945, 3475.1781349708936, 1801.224196510299, 3111.7884398157184, 3731.7534590248724, 3070.412853484885, 1626.7243408153158, 1459.9909927782753, 1142.2963071100282, 814.4987710800144, 782.0393567893748, 631.8774024418913, 573.2941832172886, 474.77562667988644, 452.5573248275798, 433.9798618253633, 432.50556419302643, 392.60946881076234, 390.8438316143026, 375.36434852994364, 364.68531207215665, 356.5146876509428, 352.43097429814156, 334.90551646168905, 331.8806835110882, 321.4740204581942, 277.8360019129116, 268.0523849133082, 266.2136726981766, 229.5605006604294, 218.47494664035392, 208.55960094952405, 196.325969591654, 179.84993114542584, 175.07414056804961, 157.0555773213148, 1178.2261730329164, 1843.6485751138043, 3425.4185202701615, 2756.6659917349916, 2381.548141605798, 2013.2656403120525, 620.057078289058, 3150.5373556587224, 2676.1673508013378, 1661.1826638393763, 1649.292998538546, 716.2659799970643, 858.6184236874091, 817.7084496510198, 1426.6279695093183, 1367.0576290911708, 3506.761659392054, 3111.7884398157184, 1779.578376557061, 5093.171432260336, 2360.8191342971795, 2422.838605271855, 1695.2217991946563, 4317.495781770247, 5100.510215382445, 3621.5054404815282, 1429.6988983922472, 2490.534807627617, 5033.912884602415, 3608.204936576235, 3929.4303920301236, 3070.412853484885, 4855.630177381934, 1970.4462526633413, 1803.288858494435, 1405.2646634608473, 484.75061120272153, 327.49732195811197, 4642.179108845422, 283.899259953446, 268.38289712410153, 246.93719289181973, 231.02281451119143, 221.23395876297212, 207.67645777992345, 201.6634500660207, 1226.061212252978, 186.3166588239046, 180.64635201350816, 124.90402596406605, 119.4571614083136, 118.22743223376077, 117.86580042686957, 172.0020157743776, 114.55644619732516, 114.47537695737691, 100.91866054795179, 100.33577964596371, 93.77280553640628, 82.21702525632706, 76.53532552628644, 75.09453332855256, 75.2438881767809, 150.01015206122207, 232.6037356187111, 258.96289251497586, 2034.7334859747966, 2777.0356354735177, 1773.6553502322774, 474.4146788430586, 835.3243549608403, 2183.4725990765746, 451.1503247512584, 917.5809766721899, 972.2468911093052, 357.68744504533595, 911.9702822981701, 1345.6889635748719, 2256.252886066182, 598.9915404694113, 2360.8191342971795, 540.7256070971288, 581.1539177183457, 1322.2561349771509, 3042.2460972610766, 3941.729810803361, 5205.6407583512755, 4735.208905478115, 3124.7771167305063, 2911.677285138968, 2676.6923132594516, 1965.368979545203, 928.3460115219482, 2374.8819159250547, 1285.1892546764225, 3512.122104928232, 5027.04659121248, 4317.495781770247, 4441.233462001629, 2571.259892753577, 1455.9531711739617, 1382.8208399508542, 1202.7994088824717, 844.4055427098957, 841.6003497762604, 754.3656667698064, 723.3195637700311, 701.8509924960142, 644.9241247741239, 459.1383802894687, 360.76520390215194, 346.8658593567899, 236.42449617318798, 218.93546220449156, 202.82060792474763, 122.78468538541141, 122.3967697554487, 111.83979358699825, 109.33718250490143, 99.33963028289996, 95.83097603953505, 92.8776324236309, 92.39114110003285, 73.8773994661845, 70.46476712938241, 66.56561524176962, 66.58655133254801, 66.07954028777358, 64.97317964026986, 1886.7036996373888, 1298.5291898768048, 353.42727947580676, 263.86546650893473, 2834.4960692129025, 569.5485855924126, 611.0287751023538, 468.81197064242514, 520.0516479699238, 2129.457158473019, 5799.707309531668, 556.7321499952257, 835.3167935352922, 5100.510215382445, 2422.838605271855, 2723.5038824864155, 1329.8713873356487, 2313.663804493153, 2490.534807627617, 5093.171432260336, 1527.0116960741154, 4441.233462001629, 2407.595548380528, 3512.122104928232, 3731.7534590248724, 4592.672118407945, 4735.208905478115, 5205.6407583512755, 5033.912884602415, 1738.5849284086787, 1699.387007110831, 1464.1678002874796, 1414.579723753747, 1239.9791848308514, 471.8734036428063, 445.9760317359506, 428.2849994487182, 336.95961437104745, 285.080531555053, 280.7288434007416, 248.6911270805866, 233.75873492349405, 230.23216473557335, 225.1251356889685, 163.8825962931115, 146.63998069990183, 141.5440793903268, 133.88649368575793, 121.70929810953642, 110.02306416852056, 106.09201618848938, 105.87116301648531, 99.47476598276131, 99.10592436400758, 87.38446446427199, 85.65317433148296, 80.7370666663746, 74.96496437374662, 74.78981057645899, 4089.597820378791, 896.1780669370394, 562.8048380145536, 473.78515897171513, 1479.434662384963, 370.6445552687202, 2578.2747642739077, 318.7606080081427, 743.592250096933, 181.38294714978548, 5033.912884602415, 1276.7759158186243, 708.1686342926497, 1050.9959980691747, 926.6986075832971, 1257.2142036244145, 3615.9997022256066, 1508.48212998749, 967.782989873525, 3042.2460972610766, 4855.630177381934, 1791.3459534612293, 1191.8468570191887, 4317.495781770247, 2676.6923132594516, 3124.7771167305063, 4536.1907143594035, 2395.9552754614315, 1706.293313510841, 2016.8223830738018, 3066.2905724927878, 3512.122104928232, 3929.4303920301236, 4592.672118407945, 5027.04659121248, 2098.633958648689, 1219.4010033368695, 1028.0643269347063, 811.036130623194, 806.9090175908167, 499.68074289693516, 442.9151784874534, 391.3277693829598, 342.7543564652983, 319.3410597530985, 288.3848864559483, 282.37412619165644, 269.4825972130807, 268.76248839396055, 266.7337258686658, 247.50254493945965, 243.4931575719467, 241.65640993849925, 231.83401396024905, 223.19639555471088, 222.0942162888185, 210.4542653385214, 200.9052466635038, 196.27379650791445, 195.67932593593025, 189.26732514618533, 188.98940594422328, 186.06266729865007, 185.12532064000823, 179.44831141575457, 819.1229826469065, 1332.0682340452872, 1474.7264211978616, 406.3835670736832, 1308.159569486324, 1634.9250199760927, 2313.406415550718, 2129.169682156315, 464.8011278831865, 919.6773435044081, 1811.1647616976245, 2080.520189747146, 553.4736370133794, 1221.8537356484105, 467.61413064669443, 3933.645335843669, 492.6727870174136, 582.4121760209694, 1449.4959575851897, 3854.6546859785094, 1512.2731349089952, 1607.4614670046399, 2911.677285138968, 4534.119526293722, 2374.8819159250547, 4878.551036994624, 3010.239248580758, 4038.207044694922, 1669.9081902312032, 3860.0724674762832, 2062.0638187097557, 2777.0356354735177, 1668.0080777959222, 1475.1637711929993, 1235.1110562659908, 598.1415716006876, 543.3247822505415, 411.81327879960264, 349.82568859693555, 303.0858224934834, 252.8612312574847, 231.64600815445112, 217.489086069143, 210.55687038385622, 194.20989304841748, 186.5359096674761, 163.9577880877368, 132.2880613351656, 130.0345884756644, 126.41414552551011, 121.11048255394374, 120.44500950445334, 114.7117265955926, 105.88966516991557, 102.32982348145826, 100.53656730973255, 89.96986417001473, 88.23481673908965, 88.1360210335684, 86.0521775675873, 84.22659327892623, 78.75686003941846, 1722.178748523132, 216.6052450680663, 861.291081080343, 2106.011448262098, 493.78662169386195, 2062.0638187097557, 3621.5054404815282, 4448.207825397681, 4534.119526293722, 870.9858701459349, 983.7383542593591, 277.8762106166183, 3854.6546859785094, 4038.207044694922, 3933.645335843669, 3668.322417739917, 2454.9833928905027, 2410.06772054874, 477.3584797432435, 1875.2295571043292, 640.3822663426602, 1459.2403791423199, 1397.2612060698928, 3258.029387749762, 625.4568301112583, 5093.171432260336, 5100.510215382445, 3941.729810803361, 1811.1647616976245, 2080.520189747146, 3731.7534590248724, 972.2351061684282, 827.0281928411766, 679.9986666021937, 533.5715099270136, 503.4763276157498, 336.7160854714374, 311.28010872033894, 191.5562884276481, 191.1137186105621, 190.55776505357056, 173.52254962159256, 169.70846295586307, 167.73339900217871, 164.4304246155927, 155.78528595225634, 148.84399879128716, 144.10531352055943, 138.94535346355016, 129.24531644126074, 124.03927599685244, 122.38312407001912, 119.57110853725332, 112.86724649913599, 102.66515605387971, 102.4575219306315, 98.4891256494589, 96.96271290296576, 95.09251701470599, 94.74017936260485, 94.13882623695302, 366.80605617839865, 1117.8480778894152, 1328.2820956781202, 463.60826649465275, 3398.5082974556017, 635.6143740160121, 420.80158854402356, 517.6757835657479, 807.4312882684283, 3608.204936576235, 273.08538112014895, 546.9921689180708, 797.5837791841262, 5205.6407583512755, 5799.707309531668, 3066.2905724927878, 2064.7821237072285, 1682.2492420434928, 3860.0724674762832, 4441.233462001629, 779.9824199554856, 1270.6405942116298, 357.3887175864178, 649.2976503014162, 3475.1781349708936, 2395.9552754614315, 2140.287364034668, 1556.0985195875662, 1193.4657603138353, 5027.04659121248, 4878.551036994624, 2834.4960692129025, 4735.208905478115, 3512.122104928232, 3615.9997022256066, 2407.595548380528, 1806.6244236333318, 3124.7771167305063, 2723.5038824864155], \"Category\": [\"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Default\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic1\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic2\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic3\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic4\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic5\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic6\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic7\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\", \"Topic8\"], \"logprob\": [30.0, 29.0, 28.0, 27.0, 26.0, 25.0, 24.0, 23.0, 22.0, 21.0, 20.0, 19.0, 18.0, 17.0, 16.0, 15.0, 14.0, 13.0, 12.0, 11.0, 10.0, 9.0, 8.0, 7.0, 6.0, 5.0, 4.0, 3.0, 2.0, 1.0, -2.8715, -3.1239, -3.5201, -4.5513, -4.7968, -5.1206, -5.3233, -5.3728, -5.3833, -5.3873, -5.5479, -5.7802, -5.8148, -5.8243, -5.9093, -5.9793, -6.0377, -6.0519, -6.113, -6.1529, -6.3031, -6.368, -6.4687, -6.5586, -6.5873, -6.6, -6.6774, -6.6986, -6.6989, -6.7806, -4.4155, -5.2485, -3.3305, -3.482, -4.9246, -5.9067, -3.7952, -4.6729, -4.0955, -3.8917, -4.4053, -5.0695, -3.8525, -4.6362, -4.2104, -4.0166, -4.9411, -4.5238, -5.0895, -5.1699, -4.6518, -4.7668, -5.1387, -5.0628, -5.0501, -5.134, -4.412, -4.5202, -4.7658, -5.1044, -5.1451, -5.3586, -5.456, -5.6449, -5.6929, -5.7349, -5.7383, -5.8353, -5.8398, -5.8803, -5.9093, -5.932, -5.9435, -5.9947, -6.0038, -6.0357, -6.1821, -6.218, -6.2249, -6.3736, -6.4233, -6.47, -6.5307, -6.6188, -6.6458, -6.755, -4.7408, -4.3003, -3.7187, -3.9695, -4.2118, -4.3735, -5.464, -4.026, -4.1877, -4.6514, -4.6883, -5.3789, -5.2383, -5.2994, -4.8953, -4.9344, -4.3114, -4.3961, -4.8036, -4.1559, -4.7373, -4.7295, -4.9583, -4.4659, -4.4073, -4.6522, -5.0547, -4.8658, -4.6972, -4.8506, -4.8666, -4.9543, -4.9424, -3.8699, -3.9586, -4.2081, -5.2737, -5.6668, -3.0153, -5.81, -5.8665, -5.95, -6.0169, -6.0604, -6.1239, -6.1534, -4.3485, -6.2329, -6.264, -6.6353, -6.6802, -6.6906, -6.6937, -6.316, -6.7224, -6.7231, -6.8503, -6.8561, -6.9244, -7.0573, -7.1298, -7.149, -7.1471, -6.4655, -6.0523, -5.9745, -4.0658, -3.7959, -4.2244, -5.4396, -4.9398, -4.1488, -5.5291, -4.934, -4.8858, -5.7563, -4.9936, -4.6879, -4.3323, -5.3766, -4.3774, -5.4633, -5.4235, -4.8942, -4.3511, -4.1917, -4.0665, -4.1641, -4.5106, -4.5581, -4.6199, -4.8824, -5.2274, -4.9199, -5.155, -4.8697, -4.8984, -4.9453, -5.0089, -3.5368, -4.1058, -4.1574, -4.2969, -4.6511, -4.6544, -4.7639, -4.806, -4.8362, -4.9209, -5.2613, -5.503, -5.5424, -5.927, -6.0042, -6.081, -6.5859, -6.5892, -6.6801, -6.7029, -6.7997, -6.836, -6.8676, -6.873, -7.0992, -7.1471, -7.2049, -7.2046, -7.2123, -7.2294, -3.9198, -4.3582, -5.6325, -5.9193, -3.758, -5.2449, -5.1929, -5.4406, -5.355, -4.1441, -3.403, -5.3707, -5.0443, -3.6916, -4.2775, -4.217, -4.7578, -4.4305, -4.3916, -3.9775, -4.7718, -4.2174, -4.5502, -4.3688, -4.3927, -4.4419, -4.4865, -4.5477, -4.8264, -4.1312, -4.1541, -4.3031, -4.3376, -4.4694, -5.4368, -5.4933, -5.5339, -5.7743, -5.942, -5.9574, -6.079, -6.1412, -6.1564, -6.1789, -6.498, -6.6098, -6.6454, -6.7014, -6.7974, -6.8992, -6.9359, -6.938, -7.0009, -7.0046, -7.1317, -7.1519, -7.2117, -7.2868, -7.2892, -3.3259, -4.8372, -5.3147, -5.4934, -4.4528, -5.7603, -3.9479, -5.911, -5.16, -6.4531, -3.552, -4.7485, -5.2635, -4.9329, -5.062, -4.8489, -4.0426, -4.8691, -5.1971, -4.4868, -4.2598, -4.8667, -5.1487, -4.4624, -4.7262, -4.6853, -4.5157, -4.8383, -4.9998, -4.9463, -4.811, -4.9224, -4.903, -4.8795, -5.0136, -4.229, -4.7723, -4.9431, -5.1804, -5.1855, -5.6655, -5.7863, -5.9104, -6.0433, -6.1143, -6.2165, -6.2377, -6.2846, -6.2872, -6.2948, -6.3699, -6.3863, -6.3939, -6.4356, -6.4737, -6.4787, -6.5328, -6.5794, -6.6028, -6.6059, -6.6394, -6.6408, -6.6565, -6.6616, -6.6929, -5.2141, -4.7594, -4.6742, -5.9324, -4.8604, -4.66, -4.3984, -4.4751, -5.8314, -5.2363, -4.7014, -4.6839, -5.7408, -5.1255, -5.9142, -4.4607, -5.8798, -5.7697, -5.1838, -4.5642, -5.2185, -5.2159, -4.9073, -4.6724, -5.0336, -4.6902, -5.0638, -4.9745, -5.3928, -5.2687, -5.437, -5.4167, -3.9662, -4.0891, -4.2668, -4.9927, -5.0891, -5.3667, -5.5302, -5.6741, -5.8558, -5.9438, -6.0071, -6.0397, -6.1209, -6.1614, -6.2911, -6.5071, -6.5244, -6.5528, -6.596, -6.6016, -6.6507, -6.7314, -6.7659, -6.7838, -6.8959, -6.9156, -6.9167, -6.9409, -6.9626, -7.0305, -3.981, -6.0696, -4.7943, -4.0146, -5.3482, -4.1164, -3.6262, -3.4847, -3.5102, -4.9235, -4.8313, -5.9041, -3.7401, -3.8608, -3.9568, -4.0837, -4.3981, -4.4335, -5.6001, -4.8252, -5.4475, -5.0378, -5.1193, -4.8359, -5.4949, -4.72, -4.7329, -4.9353, -5.1656, -5.1611, -5.1811, -4.7691, -4.931, -5.127, -5.3699, -5.428, -5.8313, -5.91, -6.3975, -6.3998, -6.4027, -6.4969, -6.5192, -6.531, -6.551, -6.6053, -6.6512, -6.6838, -6.7205, -6.7933, -6.8348, -6.8483, -6.8717, -6.9299, -7.0255, -7.0276, -7.0674, -7.0832, -7.1029, -7.1066, -7.113, -5.8043, -4.7629, -4.6117, -5.6039, -3.7826, -5.3514, -5.7318, -5.5521, -5.1549, -3.9092, -6.1558, -5.5761, -5.2668, -3.7647, -3.6889, -4.2498, -4.5827, -4.7541, -4.1416, -4.0697, -5.3789, -5.0752, -5.9756, -5.5758, -4.5528, -4.8203, -4.9275, -5.0947, -5.2479, -4.6303, -4.7884, -4.9968, -4.839, -4.9642, -5.032, -5.1403, -5.1995, -5.1843, -5.202], \"loglift\": [30.0, 29.0, 28.0, 27.0, 26.0, 25.0, 24.0, 23.0, 22.0, 21.0, 20.0, 19.0, 18.0, 17.0, 16.0, 15.0, 14.0, 13.0, 12.0, 11.0, 10.0, 9.0, 8.0, 7.0, 6.0, 5.0, 4.0, 3.0, 2.0, 1.0, 2.4217, 2.4217, 2.4216, 2.4208, 2.4205, 2.4199, 2.4195, 2.4194, 2.4193, 2.4193, 2.4189, 2.4181, 2.4179, 2.4179, 2.4175, 2.4172, 2.4169, 2.4168, 2.4165, 2.4163, 2.4154, 2.4149, 2.4142, 2.4135, 2.4132, 2.4131, 2.4124, 2.4122, 2.4122, 2.4114, 2.2573, 2.2138, 1.8985, 1.8234, 2.0929, 2.2665, 1.6143, 1.7942, 1.5708, 1.4809, 1.6512, 1.8842, 1.2737, 1.5886, 1.276, 1.1443, 1.6902, 1.0361, 1.3828, 1.4715, 0.5649, 0.7286, 1.0139, 0.5431, 0.3741, 0.4852, 1.8425, 1.8424, 1.8422, 1.8419, 1.8419, 1.8416, 1.8414, 1.8411, 1.841, 1.8409, 1.8409, 1.8407, 1.8407, 1.8406, 1.8405, 1.8405, 1.8404, 1.8403, 1.8403, 1.8402, 1.8397, 1.8396, 1.8396, 1.839, 1.8388, 1.8386, 1.8384, 1.8379, 1.8378, 1.8372, 1.8363, 1.829, 1.7911, 1.7575, 1.6615, 1.6678, 1.7551, 1.5675, 1.5689, 1.5822, 1.5524, 1.6959, 1.6552, 1.6429, 1.4905, 1.494, 1.175, 1.2097, 1.3611, 0.9572, 1.1448, 1.1267, 1.2549, 0.8125, 0.7044, 0.8019, 1.3289, 0.9627, 0.4277, 0.6072, 0.5059, 0.665, 0.2185, 2.193, 2.1929, 2.1928, 2.1915, 2.1906, 2.1906, 2.1902, 2.1899, 2.1897, 2.1894, 2.1892, 2.189, 2.1888, 2.1887, 2.1885, 2.1883, 2.186, 2.1857, 2.1856, 2.1856, 2.1853, 2.1853, 2.1853, 2.1842, 2.1842, 2.1835, 2.1821, 2.1813, 2.1811, 2.181, 2.1726, 2.1472, 2.1176, 1.9649, 1.9238, 1.9436, 2.0472, 1.9812, 1.8114, 2.008, 1.8931, 1.8834, 2.0129, 1.8396, 1.7562, 1.5951, 1.877, 1.5047, 1.8926, 1.8603, 1.5675, 1.2774, 1.1777, 1.0249, 1.0219, 1.0911, 1.1142, 1.1365, 1.183, 1.588, 0.9563, 1.3351, 0.6151, 0.2279, 0.3331, 0.2412, 2.2599, 2.2596, 2.2596, 2.2595, 2.2591, 2.2591, 2.259, 2.259, 2.2589, 2.2588, 2.2582, 2.2576, 2.2575, 2.2562, 2.2559, 2.2555, 2.2524, 2.2524, 2.2517, 2.2515, 2.2506, 2.2502, 2.2499, 2.2498, 2.2472, 2.2466, 2.2458, 2.2458, 2.2457, 2.2454, 2.1865, 2.1216, 2.1487, 2.1541, 1.9412, 2.0591, 2.0407, 2.058, 2.0399, 1.8411, 1.5803, 1.956, 1.8767, 1.4201, 1.5786, 1.5221, 1.6982, 1.4717, 1.437, 1.1357, 1.5459, 1.0328, 1.3122, 1.116, 1.0315, 0.7747, 0.6995, 0.5436, 0.2985, 2.0568, 2.0567, 2.0567, 2.0566, 2.0565, 2.0553, 2.0552, 2.0551, 2.0546, 2.0541, 2.054, 2.0536, 2.0534, 2.0533, 2.0532, 2.0517, 2.051, 2.0508, 2.0505, 2.0498, 2.049, 2.0487, 2.0486, 2.0481, 2.048, 2.0468, 2.0466, 2.0459, 2.045, 2.045, 2.0067, 2.0134, 2.0012, 1.9947, 1.8966, 1.9733, 1.8461, 1.9734, 1.8773, 1.9951, 1.5728, 1.7483, 1.8227, 1.7585, 1.7552, 1.6633, 1.4131, 1.4609, 1.5768, 1.1416, 0.9012, 1.2914, 1.4168, 0.816, 1.0303, 0.9164, 0.7133, 1.029, 1.207, 1.0932, 0.8096, 0.5625, 0.4696, 0.3371, 0.1127, 1.7708, 1.7705, 1.7703, 1.7701, 1.7701, 1.7694, 1.7691, 1.7688, 1.7685, 1.7683, 1.768, 1.7679, 1.7678, 1.7678, 1.7677, 1.7675, 1.7674, 1.7674, 1.7672, 1.7671, 1.767, 1.7668, 1.7666, 1.7665, 1.7665, 1.7663, 1.7663, 1.7662, 1.7662, 1.766, 1.7265, 1.695, 1.6784, 1.7091, 1.612, 1.5895, 1.5039, 1.5103, 1.6759, 1.5885, 1.4457, 1.3246, 1.5918, 1.4152, 1.587, 0.9108, 1.5692, 1.5119, 1.1861, 0.8276, 1.109, 1.0506, 0.765, 0.5571, 0.8426, 0.466, 0.5752, 0.3708, 0.8355, 0.1217, 0.5803, 0.303, 2.2633, 2.2632, 2.2631, 2.2623, 2.2621, 2.2616, 2.2612, 2.2608, 2.2602, 2.2598, 2.2595, 2.2594, 2.259, 2.2588, 2.2581, 2.2568, 2.2567, 2.2565, 2.2561, 2.256, 2.2557, 2.255, 2.2547, 2.2546, 2.2534, 2.2532, 2.2532, 2.253, 2.2527, 2.252, 2.2165, 2.2012, 2.0961, 1.9817, 2.0985, 1.9009, 1.828, 1.7639, 1.7192, 1.9557, 1.9261, 2.1176, 1.6517, 1.4844, 1.4147, 1.3577, 1.4449, 1.4279, 1.8804, 1.2872, 1.7393, 1.3254, 1.2872, 0.7241, 1.7155, 0.3932, 0.3788, 0.4341, 0.9815, 0.8473, 0.2431, 2.0002, 2.0, 1.9998, 1.9994, 1.9993, 1.9983, 1.9981, 1.9962, 1.9962, 1.9962, 1.9957, 1.9955, 1.9955, 1.9954, 1.995, 1.9948, 1.9945, 1.9943, 1.9938, 1.9935, 1.9934, 1.9932, 1.9927, 1.9919, 1.9918, 1.9915, 1.9913, 1.9911, 1.9911, 1.991, 1.9397, 1.8668, 1.8455, 1.9059, 1.7351, 1.8429, 1.8749, 1.8474, 1.8, 1.5486, 1.8832, 1.7683, 1.7004, 1.3266, 1.2944, 1.3708, 1.4334, 1.4669, 1.2488, 1.1804, 1.6107, 1.4263, 1.7945, 1.5972, 0.9426, 1.047, 1.0526, 1.2042, 1.3163, 0.4959, 0.3678, 0.7024, 0.347, 0.5207, 0.4237, 0.7222, 0.9502, 0.4174, 0.5371]}, \"token.table\": {\"Topic\": [1, 2, 5, 8, 1, 4, 2, 5, 4, 8, 3, 8, 6, 5, 6, 8, 3, 1, 1, 2, 5, 8, 5, 5, 5, 7, 5, 7, 3, 4, 7, 8, 8, 2, 3, 4, 6, 7, 2, 3, 5, 6, 3, 4, 5, 6, 8, 3, 4, 5, 8, 2, 3, 8, 3, 4, 5, 6, 8, 2, 4, 5, 6, 5, 2, 1, 6, 2, 3, 7, 2, 4, 6, 7, 7, 8, 1, 2, 5, 6, 7, 6, 7, 1, 2, 3, 4, 5, 6, 7, 8, 3, 6, 7, 5, 2, 6, 2, 3, 2, 3, 5, 6, 8, 6, 7, 3, 6, 1, 2, 3, 4, 6, 8, 1, 1, 3, 2, 4, 1, 5, 4, 8, 4, 7, 1, 2, 4, 6, 7, 8, 4, 2, 2, 7, 2, 3, 8, 4, 6, 8, 8, 8, 7, 8, 8, 1, 4, 7, 1, 2, 3, 4, 5, 7, 8, 1, 1, 3, 4, 8, 5, 6, 5, 5, 6, 3, 8, 1, 2, 4, 6, 8, 6, 1, 2, 3, 6, 6, 7, 1, 2, 3, 5, 6, 7, 8, 8, 5, 6, 4, 5, 3, 6, 6, 4, 4, 2, 3, 4, 5, 6, 7, 2, 4, 7, 6, 6, 4, 2, 4, 7, 3, 3, 2, 3, 8, 1, 2, 3, 4, 5, 6, 7, 8, 1, 1, 2, 3, 7, 8, 4, 8, 1, 2, 3, 5, 8, 8, 2, 1, 3, 4, 5, 6, 8, 4, 3, 6, 8, 3, 6, 6, 7, 2, 3, 5, 6, 8, 6, 6, 2, 3, 7, 8, 1, 2, 5, 6, 7, 7, 4, 8, 5, 6, 6, 1, 2, 5, 7, 7, 5, 6, 7, 7, 2, 3, 4, 5, 2, 3, 6, 7, 1, 6, 3, 8, 1, 2, 3, 5, 7, 8, 2, 3, 4, 6, 7, 8, 3, 5, 8, 8, 6, 8, 2, 3, 5, 6, 8, 3, 6, 8, 1, 2, 3, 5, 5, 2, 3, 4, 5, 7, 8, 4, 7, 2, 4, 5, 6, 7, 8, 2, 1, 2, 3, 5, 8, 4, 5, 3, 8, 2, 5, 2, 5, 6, 8, 2, 3, 4, 5, 6, 7, 8, 6, 7, 4, 1, 2, 3, 4, 5, 8, 6, 1, 4, 5, 1, 6, 7, 2, 3, 6, 4, 7, 1, 2, 6, 3, 4, 5, 7, 8, 2, 4, 2, 1, 4, 3, 8, 3, 5, 1, 3, 6, 1, 2, 3, 5, 6, 7, 8, 1, 2, 5, 7, 3, 5, 6, 2, 6, 3, 4, 6, 8, 3, 8, 1, 6, 7, 1, 2, 5, 7, 8, 6, 3, 5, 6, 8, 8, 2, 4, 7, 8, 2, 8, 1, 2, 5, 8, 5, 3, 8, 3, 4, 8, 6, 7, 4, 8, 2, 3, 2, 4, 8, 5, 8, 4, 1, 2, 5, 8, 4, 3, 3, 4, 6, 2, 6, 4, 6, 7, 8, 3, 4, 5, 8, 8, 2, 4, 6, 7, 1, 2, 4, 6, 7, 3, 5, 6, 7, 2, 3, 4, 5, 6, 8, 5, 5, 6, 6, 1, 6, 7, 2, 5, 6, 7, 8, 1, 2, 3, 4, 5, 6, 7, 8, 6, 7, 2, 3, 6, 1, 1, 2, 3, 4, 5, 6, 7, 8, 7, 1, 2, 5, 8, 1, 2, 3, 5, 1, 1, 5, 6, 3, 6, 8, 2, 3, 4, 6, 7, 2, 3, 3, 3, 4, 4, 1, 8, 2, 3, 7, 1, 1, 3, 8, 5, 8, 2, 3, 5, 6, 8, 1, 5, 6, 4, 7, 7, 2, 2, 6, 7, 2, 3, 4, 6, 1, 6, 8, 4, 6, 3, 4, 1, 2, 3, 5, 6, 7, 8, 7, 2, 3, 7, 2, 3, 4, 7, 2, 3, 4, 7, 2, 3, 4, 5, 7, 8, 3, 5, 1, 2, 5, 6, 7, 7, 7, 4, 7, 7, 6, 7, 3, 1, 2, 4, 5, 6, 7, 2, 2, 6, 4, 4, 8, 2, 4, 5, 6, 7, 8, 1, 4, 5, 6, 7, 8, 2, 2, 2, 4, 7, 8, 3, 8, 2, 4, 7, 3, 6, 4, 6, 7, 5, 5, 2, 3, 6, 4, 5, 5, 5, 4, 5, 6, 1, 2, 3, 4, 7, 8, 4, 6, 4, 8, 3, 6, 8, 2, 3, 5, 4, 5, 5, 5, 1, 2, 3, 4, 5, 7, 1, 2, 3, 5, 8, 5, 1, 2, 5, 6, 7, 2, 8, 3, 1, 1, 5, 3, 2, 1, 6, 7, 8, 5, 6, 6, 6, 2, 3, 4, 2, 3, 5, 6, 8, 6, 7, 7, 1, 2, 3, 5, 7, 2, 5, 2, 3, 5, 6, 7, 7, 7, 2, 4, 6, 7, 8, 5, 5, 2, 3, 6, 6, 6, 7, 5, 5, 8, 6, 7, 1, 2, 4, 5, 6, 7, 1, 2, 3, 4, 5, 6, 7, 8, 1, 2, 8, 1, 1, 2, 4, 6, 7, 8, 2, 3, 1, 6, 6, 4, 4, 2, 7, 1, 6, 4, 1, 1, 2, 5, 2, 3, 5, 3, 6, 6, 7, 1, 2, 3, 4, 5, 8, 2, 3, 4, 6, 7, 8, 6, 7, 3, 8, 1, 2, 3, 5, 6, 7, 8, 1, 5, 3, 3, 6, 3, 2, 3, 5, 2, 3, 7, 2, 6, 2, 3, 4, 5, 6, 7, 4, 4, 1, 2, 2, 3, 4, 6, 7, 3, 4, 6, 5, 3, 4, 5, 6, 2, 1, 1, 2, 3, 4, 5, 8, 3, 5, 6, 8, 2, 3, 1, 2, 3, 4, 5, 6, 7, 8, 3, 6, 7, 8, 6, 6, 3, 6, 3, 4, 8, 5, 6, 7, 1, 7, 1, 2, 3, 6, 7, 1, 2, 3, 5, 8, 5, 6, 7, 1, 6, 7], \"Freq\": [0.46241049495301695, 0.08622921988764003, 0.20120151307116005, 0.24961089967474742, 0.1813093635035008, 0.8169586614334212, 0.14802951800991, 0.851676678961126, 0.7267605774355621, 0.2727116147367425, 0.6457658667954516, 0.35372215488450515, 0.9996979184263762, 0.3576861424656403, 0.2570999577115921, 0.385232566506168, 0.9908970200063715, 0.9962269605872611, 0.5925676783145563, 0.11639722252607354, 0.26079150425822917, 0.030201552056954687, 0.9952272721823383, 0.9946144599055965, 0.9956357011447543, 0.9946629636947968, 0.9992103215579502, 0.9937701780819381, 0.9967101124878927, 0.8035628107984804, 0.0016365841360457848, 0.19475351218944842, 0.9955345208237492, 0.41231677117689824, 0.09718895320598316, 0.3247485426317094, 0.01158413785687476, 0.1539315945727087, 0.9978410005581582, 0.5495837845384267, 0.25174483033695677, 0.1981160900715135, 0.34744374442780984, 0.05828088616208423, 0.35790441630305575, 0.23611230804126432, 0.9987567620426042, 0.0692564053131009, 0.29040469955152537, 0.07712645137140782, 0.5627082931689448, 0.9181380724354778, 0.0816203343729682, 0.9989288972211208, 0.07684014872200627, 0.3875235608520641, 0.16032593192807795, 0.09677705217420249, 0.27828594402023893, 0.7592355620553735, 0.14568943268537474, 0.09331741439978251, 0.0019044370285669903, 0.9897068956956268, 0.9960739580300749, 0.8480560729797113, 0.15130515839505862, 0.7051641273095953, 0.10606721831938934, 0.18799500074540043, 0.9969910238156476, 0.06848862518355084, 0.3891399158156298, 0.5422016160364441, 0.9860053345751144, 0.9916213958159606, 0.00044109997286172984, 0.06440059603781256, 0.058004646431317475, 0.2968602817359442, 0.5800464643131747, 0.24548774434566045, 0.7540319884350267, 0.09438844397979619, 0.06452543345511491, 0.06612523759036569, 0.08638942330354228, 0.07572406240187039, 0.18184440337350566, 0.3764872398290176, 0.054926608643610215, 0.9951426024829133, 0.3040643040778036, 0.6959047469723256, 0.9871277952066291, 0.8181400110094431, 0.18099360482720117, 0.9932789567914853, 0.9834685818752685, 0.09728690764100877, 0.13538117870525918, 0.42724190162828546, 0.16234020130457485, 0.17757790973027504, 0.13315281620508476, 0.863693942951901, 0.9950563542893262, 0.004078099812661173, 0.059021995974945646, 0.03765196294953429, 0.3643081820522507, 0.08140964962061469, 0.19538315908947523, 0.26203730971635353, 0.9893552475079047, 0.9968876937949764, 0.9930055105586167, 0.9977425495405766, 0.9885881812221068, 0.854284941380821, 0.14238082356347015, 0.9924911021375858, 0.9969885780105692, 0.9911911441453298, 0.9963052889755728, 0.11018569827753508, 0.1736622418504629, 0.05629051977221901, 0.39223713245535585, 0.24672015049100246, 0.020360400768674958, 0.9995100095649111, 0.9975583749869125, 0.42049265007341385, 0.5771781242452563, 0.9986709661344388, 0.9799336776887368, 0.013332430988962406, 0.9994064018077111, 0.18082033991165167, 0.8174070160389733, 0.9943296417850838, 0.9886984085385188, 0.9871105931462947, 0.9970729870104544, 0.9956275911265043, 0.18084497548679743, 0.8181918308431806, 0.9975617120848896, 0.12889383108542438, 0.1733769411897496, 0.04287528684754241, 0.29262383273447695, 0.17471679390373532, 0.13264541868458432, 0.05493396127341371, 0.9960301135124275, 0.5337707492119675, 0.11253286450622677, 0.35356609456348276, 0.9885110096040781, 0.9507047320462142, 0.0491490872276978, 0.9995901795112625, 0.9392442789170743, 0.05909851642624288, 0.9941743951670194, 0.005813885351853915, 0.9991106996605397, 0.9932460815326065, 0.16173849489762468, 0.09779536900786608, 0.7397342014697562, 0.9971629657155024, 0.021821454893454717, 0.20631193717448096, 0.25524489663253097, 0.5157798429362024, 0.7709316846147702, 0.22664361329431998, 0.18387546628752946, 0.034818359030971927, 0.07740610396141694, 0.20401831861949668, 0.11826731869197904, 0.03453060399765811, 0.34703257017646405, 0.9987296219190277, 0.9573990166178419, 0.04240228744927505, 0.19004956600814168, 0.8094560086918199, 0.9854245937748388, 0.9951977842078773, 0.998851072507702, 0.9905506589613755, 0.9877701027750418, 0.2139463819413634, 0.09508728086282818, 0.4546900884895238, 0.170292675727065, 0.003889934217115698, 0.06223894747385117, 0.8292391362186896, 0.08618496640475146, 0.08385564298840682, 0.9954941611603604, 0.9964025384109855, 0.9881235740223083, 0.10683794708628898, 0.20948617075742937, 0.6808300549616455, 0.9929332200769604, 0.9866869061995928, 0.9963652687441231, 0.8642228377078365, 0.1349030771056135, 0.08691105141357718, 0.1184778012194519, 0.08014674788374687, 0.054524386028329076, 0.13364623943785922, 0.271187077877742, 0.06005881618909932, 0.1953448867857053, 0.9858943297158405, 0.35389869341034413, 0.175038446632762, 0.08407960318604289, 0.016815920637208576, 0.3699502540185887, 0.18467286948990208, 0.8142394700236592, 0.27864560326328486, 0.19709079255207954, 0.11821328623290873, 0.3146862393099034, 0.09123429582086859, 0.9879026934743532, 0.9973463851473032, 0.0041331579929119655, 0.0960959233352032, 0.06303065939190748, 0.6189404094385669, 0.0030998684946839743, 0.21389092613319424, 0.9836630175834724, 0.13551338272620894, 0.008281373388823881, 0.8559928675538866, 0.9544128747953861, 0.0429915709367291, 0.8179871318643704, 0.1806472822231488, 0.06287847210071687, 0.739746730596669, 0.14609997929284213, 0.04993290431527515, 0.9935211119385955, 0.9972837056601713, 0.9979695362745358, 0.9493730417921358, 0.05021284230880915, 0.9892201218823451, 0.9913007302697371, 0.9980717185858662, 0.26874672200074146, 0.04354692254641644, 0.4864813347328237, 0.20093794260703587, 0.9964174421470777, 0.08843673196339111, 0.9102512411841719, 0.9191850957710413, 0.07842876243780215, 0.9951336681933678, 0.31795716626869386, 0.5127237533193826, 0.08184188943418398, 0.08725999361275365, 0.9908308304076889, 0.7338797579050731, 0.03289535734473113, 0.23183394700096227, 0.9965940557466965, 0.2427535056750421, 0.0003973052466039969, 0.1406460572978149, 0.6160217848594972, 0.2665478835164394, 0.682399220686417, 0.05037846595671535, 0.9903899160144317, 0.04272862652064934, 0.9559004161619553, 0.9926543541575764, 0.9958254117471601, 0.446006421871207, 0.2305290741820981, 0.13467334230569306, 0.07605082859615607, 0.00026406537706998636, 0.11249185063181419, 0.08628590600188805, 0.02313196628986786, 0.4780606366572691, 0.014319788655632484, 0.16706420098237898, 0.23131966289867859, 0.07238225573466853, 0.6745071743186694, 0.252940190369501, 0.9903647073986878, 0.05725096313509749, 0.9405515372194587, 0.06457313660232561, 0.05902897840919665, 0.2873178451850953, 0.05674608974143766, 0.5322391865403808, 0.741079008052423, 0.1863595740837711, 0.07192825666391164, 0.3865098767734349, 0.268837080566066, 0.05793122274824317, 0.2860354123194507, 0.9894395965122609, 0.1133226106368591, 0.17193775406971726, 0.055823946126531576, 0.465013471234008, 0.061964580200450044, 0.13230275231987984, 0.9959540209787158, 0.997355249520754, 0.3116403952143298, 0.060272280333805904, 0.11353615597763439, 0.028033618759909725, 0.09858489263901586, 0.387331165866086, 0.9978260819024903, 0.42704693759732715, 0.16180675333423908, 0.08056201220860848, 0.22052144019814016, 0.11026072009907008, 0.2075776769566053, 0.7907721026918297, 0.9962047317342337, 0.9949591776434071, 0.07824207745065828, 0.920018910712913, 0.29072628036349246, 0.0349204111780744, 0.0382461646236053, 0.6360503464577838, 0.10065645119166387, 0.11726642663583282, 0.12656801288456743, 0.12656801288456743, 0.30230155308387496, 0.047172330261439833, 0.17938773479702472, 0.4075690626016529, 0.5918340027817105, 0.9936092568633593, 0.24483348650012346, 0.2026399604819616, 0.1959778247948834, 0.0410831700703155, 0.051631551574855966, 0.26426471558743486, 0.9958005407944243, 0.9996530662138483, 0.05330444582856309, 0.9452655060265188, 0.03287666576630358, 0.7702533122391124, 0.19679032794401716, 0.598028019019586, 0.35112288368167505, 0.0503602542332283, 0.9291337056989469, 0.07049331594863659, 0.9865516093026743, 0.059057506121177515, 0.9399986390954088, 0.20642790265824623, 0.31832606230609556, 0.22436577557889384, 0.023347707611001644, 0.22749778513646723, 0.9985563353523131, 0.9975037631019799, 0.9977421490913383, 0.994892549331912, 0.9915028916999373, 0.9958473431578748, 0.9950337090898768, 0.9984814472523363, 0.9923741958592786, 0.9955393505675765, 0.9971610081092498, 0.0025849928920524947, 0.004958294832467688, 0.03768304072675443, 0.17701112551909648, 0.38129287261676525, 0.2256024148772798, 0.04710380090844304, 0.12643651822792606, 0.8124500232703686, 0.06581873606240961, 0.12135329461506772, 0.9854369833662365, 0.733352820687848, 0.1470819822697928, 0.11828271301416902, 0.9160446995739477, 0.08225049239132277, 0.09552012616601788, 0.031002146211777732, 0.36867417116708656, 0.5044132978240593, 0.831208533888912, 0.16845826286815282, 0.9973654530163386, 0.9978415009180702, 0.9980914692191806, 0.43443408085701624, 0.2655537994249618, 0.04594975855218441, 0.049530259218588384, 0.20408853798502685, 0.9986376443226759, 0.33218369221996624, 0.31938277922497715, 0.1430502027190028, 0.20513463074469976, 0.9958875987110085, 0.19701153246473552, 0.5699799298025554, 0.03158200138747668, 0.2007712945346732, 0.2333376677625604, 0.7665127673662924, 0.11410513586329205, 0.02564160356478473, 0.1820553853099716, 0.6769383341103169, 0.9908707772426221, 0.1252406322193003, 0.8740006977018313, 0.20945709558145828, 0.12321005622438722, 0.6684145550173006, 0.9263789710325356, 0.07281909253659316, 0.5065772879902876, 0.49330075593608336, 0.9952742203457471, 0.9967427324832345, 0.10777175343747356, 0.7382365110466939, 0.15267665070308753, 0.9950021765205154, 0.9985313697640265, 0.9934042621821337, 0.583838837082359, 0.022265040397208606, 0.1731725364227336, 0.21893956390588462, 0.9850218252260707, 0.9964220035095985, 0.9992852734112098, 0.984942916783259, 0.9977991338371954, 0.9938646532002703, 0.9975017239659842, 0.9913287323798912, 0.8329008052771809, 0.16636269348225677, 0.9900713080921201, 0.1418524843132349, 0.2929366382405057, 0.12474012112624148, 0.4401930267180543, 0.9990539225190477, 0.10244950055050313, 0.046775950623554774, 0.4230172925956258, 0.42784741793175374, 0.05684487569868242, 0.08879418539794189, 0.25020043829420074, 0.17053462709604725, 0.43359777448994985, 0.11914282373084598, 0.526913333119516, 0.08809856684323118, 0.2651347344996291, 0.00034344465477130384, 0.3400102082235908, 0.0024041125833991267, 0.22976447404200226, 0.3657685573314386, 0.062163482513606, 0.9981490721111558, 0.7391831544739083, 0.2600629784353458, 0.9966070146643224, 0.07522325390648556, 0.2104218048465204, 0.713604381653417, 0.18445249564589286, 0.06865156087944045, 0.1922243704624333, 0.08367718552475195, 0.4712346763762347, 0.10304437749855391, 0.1373925033314052, 0.046342709457021604, 0.04198104268459604, 0.038164584258723676, 0.1504775036486819, 0.40399938479591774, 0.07851000190366013, 0.15325829199859875, 0.8452427013256052, 0.49728502405988084, 0.5023680055664554, 0.9947647544618703, 0.994670832273852, 0.07839197885991657, 0.1955994035630928, 0.3620237988126244, 0.002536957244657494, 0.0030443486935889928, 0.12177394774355971, 0.16033569786235363, 0.07610871733972482, 0.9976397142238262, 0.3901328811191848, 0.2626334855283749, 0.20435531868147777, 0.14276878428432008, 0.5496567329053225, 0.21962474652450764, 0.1480028086004808, 0.08256748325460905, 0.9982030433046232, 0.9938932851274229, 0.4422226889600188, 0.5567452573958428, 0.99276223518747, 0.9989647272969001, 0.9978733256225796, 0.7604158235435651, 0.05978699349728275, 0.04521391383232008, 0.0011210061280740517, 0.13339972924081214, 0.9930181715351466, 0.0059411343596119884, 0.9991000531830547, 0.7158172513630253, 0.2821971856335004, 0.99833546484621, 0.9926523834184038, 0.994172482129179, 0.7705353708914977, 0.03973073006159285, 0.1896239389303295, 0.9931594118583112, 0.06950802806220222, 0.9267737074960296, 0.9970959531936315, 0.9946481642259303, 0.9923298212012027, 0.005587976599034851, 0.40003338358973023, 0.4003620880955558, 0.0864492850321274, 0.10781507791079008, 0.9961368654577941, 0.8351351159443201, 0.16406841247215306, 0.8941018940832246, 0.10185970945251926, 0.9902632080161629, 0.9973168295922301, 0.9983663300382322, 0.9946397182994937, 0.9870045365445274, 0.488682984258107, 0.0024764340418485156, 0.5056052835440719, 0.002889173048823268, 0.9946090274760856, 0.1098557523546123, 0.8898315940723596, 0.8018434354122389, 0.19613436549651886, 0.0985350616129967, 0.8981849847030853, 0.13955873043128478, 0.10520581217127621, 0.11522541333044538, 0.07085289391126766, 0.1667547907204582, 0.3764507292659271, 0.025764688695006418, 0.998025127305333, 0.8628079753313608, 0.1354245527623657, 0.9888133917321401, 0.4147703524705982, 0.0879329208045121, 0.4388615636499165, 0.05822042701668609, 0.555679498958492, 0.09202335651541906, 0.30674452171806355, 0.04601167825770953, 0.6175620104612803, 0.11800547970597715, 0.1522832619062848, 0.02247723422970993, 0.0033715851344564896, 0.08653735178438324, 0.9851974058592761, 0.9941721945606977, 0.0609234124477473, 0.10588534045346486, 0.12162201525546601, 0.10498610189335052, 0.6065364087971299, 0.9993956397397942, 0.9991004401909009, 0.9975206161402778, 0.9971270428925378, 0.9915981869572638, 0.2640685777846926, 0.7347995207921882, 0.9944223808592867, 0.0493407399007959, 0.2494448517206904, 0.12061069753527887, 0.06715822930941664, 0.12266656169781204, 0.3912994789354786, 0.9984476461746874, 0.998831080488017, 0.9933040468279114, 0.998567079849203, 0.9993453284124564, 0.9952236912056779, 0.10835103845114448, 0.10590703006502844, 0.08839163663119681, 0.12138574984376337, 0.44073617896292605, 0.1348277959674016, 0.04815610752401591, 0.1893033192323384, 0.1721442464364247, 0.16273443232253654, 0.07804610529754302, 0.34982367764572475, 0.9988651743842998, 0.9987684989348208, 0.9954409828545786, 0.6574445484519188, 0.06010921585846115, 0.28176194933653664, 0.14101490221770932, 0.8576796792419581, 0.9993212336355655, 0.9992866569311432, 0.9977512155759967, 0.9984515518178323, 0.9942886592239142, 0.11727836201851725, 0.6397438518305183, 0.2427277574563574, 0.9974037459353574, 0.9962097322144063, 0.005761539317543475, 0.7637640557818568, 0.23046157270173898, 0.9987875026107923, 0.996156113398241, 0.999183819162423, 0.9996635606353644, 0.08753601361852638, 0.7421531589396801, 0.17031463519256762, 0.08352105613184382, 0.3203601073226122, 0.01176352903265406, 0.431721515498404, 0.15174952452123738, 0.000980294086054505, 0.8702153242371135, 0.1286070434934495, 0.9957272239267578, 0.9923162252465987, 0.5461325774091651, 0.1432655479199585, 0.3102291563981056, 0.009337832266960265, 0.7956816021162458, 0.19462008303769818, 0.9939748368031014, 0.997152138327204, 0.9992024136255078, 0.9978114704233064, 0.08750517979360796, 0.19158331030570228, 0.024527967063359805, 0.08750517979360796, 0.5508848818824864, 0.05833678652907197, 0.10364289652787777, 0.13463890296612158, 0.07942726649799979, 0.1152663989422192, 0.5666457426991448, 0.9972209419423209, 0.1526453385848203, 0.5308844196676276, 0.18703070959234822, 0.0028922274679229107, 0.1266152913735141, 0.9993876343369394, 0.9921872708328753, 0.9968324681311481, 0.9984448091891588, 0.9966059851789755, 0.9907013663339393, 0.9961730096134548, 0.9972962032060387, 0.9960924238161165, 0.8318824742571963, 0.091956160393469, 0.07484803752956778, 0.9372435649069839, 0.05513197440629317, 0.9965283714430175, 0.9979746553173635, 0.8394321972027761, 0.0715255836669821, 0.0889102741415958, 0.0007562831236304155, 0.5346921684067037, 0.2140281239874076, 0.13310582975895313, 0.11646760103908399, 0.7221871955889851, 0.27716970350586434, 0.9877727955604506, 0.0035047679611382677, 0.7030564530043365, 0.1493031151444902, 0.1191621106787011, 0.02383242213574022, 0.9985254781785521, 0.9969996627237184, 0.7475930602340354, 0.15339906264331787, 0.0842785363929691, 0.013945369331210716, 0.0006063204057048137, 0.9992110901747145, 0.9972112269078239, 0.08741503248669322, 0.14338046405041183, 0.24639647967212394, 0.4586193772389684, 0.06388974045769645, 0.9967541964849255, 0.9933787668841444, 0.13419537270567639, 0.8359253424791092, 0.03075310624505084, 0.9972492197367383, 0.14986230235674258, 0.8485446579388533, 0.9917714796771463, 0.1447418493995266, 0.8542915676515538, 0.05540032050576202, 0.9371887552224741, 0.25015121197629825, 0.2090220556513609, 0.06967401855045363, 0.1939822895325405, 0.0626145364946808, 0.2145468676950092, 0.15611826438168394, 0.19291601428475866, 0.052257159625668265, 0.22644769171122914, 0.1789807717179138, 0.035709059077539976, 0.08970812402406385, 0.06793430751336874, 0.17550525483734852, 0.0310790555441138, 0.7916018265059573, 0.9986837595729051, 0.16065820823880797, 0.025062680485254042, 0.1310970979228673, 0.05912222063188133, 0.17286823206495736, 0.4504856159016175, 0.9987771384198002, 0.9955726688147422, 0.16575683697345509, 0.8336773756266395, 0.9982091711373143, 0.9993353763922994, 0.9978789420546238, 0.35316804600186924, 0.6466923875968551, 0.9960979887894499, 0.9935101040965338, 0.9865146439634922, 0.998429618012691, 0.48113776669489494, 0.2508725785373754, 0.2670001014433495, 0.11952143848955188, 0.7017772535166349, 0.1776373672963982, 0.08820976873856488, 0.9100666384490962, 0.8365348754430514, 0.1626093710364463, 0.9960138110592683, 0.061946907756140174, 0.04756637559846478, 0.1587389511251092, 0.5251659724504919, 0.2065818754189139, 0.4282874857353138, 0.018336467279187744, 0.4891907520554731, 0.05762889716316148, 0.005238990651196498, 0.9931962211043052, 0.9939213034925864, 0.9937955201554787, 0.11644442733579106, 0.8816506641138466, 0.9994108015670807, 0.0298962232706818, 0.2901196877957713, 0.08842544911046729, 0.39496700602675394, 0.16042902910041923, 0.03579125321137962, 0.9955868365674474, 0.9888409863375512, 0.9997735271070002, 0.9917587457047315, 0.9979337387113743, 0.9948472978758327, 0.10851572285822515, 0.7295595521391445, 0.16026937529830176, 0.8339113391429942, 0.1364658535858374, 0.029392645387718828, 0.998120812521177, 0.9988734571420075, 0.3569198623207835, 0.15564578032418333, 0.13155774289305971, 0.28905644917348333, 0.05720908889891857, 0.009959477014791492, 0.9981756835621127, 0.9981896488268691, 0.9949100990840237, 0.9986114356384632, 0.017186999873479755, 0.0752954280171494, 0.08675342793280258, 0.7005748519856511, 0.1203089991143583, 0.809266479524221, 0.1460510510383949, 0.04429417121656239, 0.9841566292960685, 0.9896180420180837, 0.6811786910111484, 0.2035156018838229, 0.11372930693507749, 0.9995547243026113, 0.9997814247733229, 0.06842359153894302, 0.14466098828449372, 0.30980681724576975, 0.2099168209558931, 0.07560384497204198, 0.19133263559963695, 0.42406205779958606, 0.3594801297310252, 0.21008579010254722, 0.006224764151186585, 0.9860881431201057, 0.013560067974698923, 0.14428027146160485, 0.3077761998447327, 0.017587211419699014, 0.04820198685398988, 0.13581235485212015, 0.08500485519521189, 0.08565623339594149, 0.17587211419699011, 0.24015466311593975, 0.11413290920361494, 0.05944422354354944, 0.5861200441393976, 0.9950731887254752, 0.9987224605858214, 0.08815194339191819, 0.9113554762979851, 0.3108166842677886, 0.1796128552474551, 0.5094473712473272, 0.07491440821587364, 0.8531069343766835, 0.07185667726828697, 0.04587212568250948, 0.954024082232444, 0.141515655771077, 0.06427404286017044, 0.7791818178311889, 0.015222799624777208, 0.9941583251463221, 0.31728371143170564, 0.17763909361035307, 0.14004246573537352, 0.14302632509053062, 0.2219991360236887, 0.12319495531966626, 0.7655377749864174, 0.11109159128826046, 0.7197951436490394, 0.27948089826666, 0.992043743993099], \"Term\": [\"absolutely\", \"absolutely\", \"absolutely\", \"absolutely\", \"accessible\", \"accessible\", \"accommodate\", \"accommodate\", \"accommodation\", \"accommodation\", \"advice\", \"advice\", \"air\", \"airbnb\", \"airbnb\", \"airbnb\", \"alessandros\", \"alice\", \"amazing\", \"amazing\", \"amazing\", \"amazing\", \"ample\", \"andreas\", \"annalisa\", \"annoying\", \"answer\", \"antonella\", \"antonio\", \"appartment\", \"appartment\", \"appartment\", \"appreciable\", \"area\", \"area\", \"area\", \"area\", \"area\", \"arno\", \"arrival\", \"arrival\", \"arrival\", \"arrive\", \"arrive\", \"arrive\", \"arrive\", \"attention\", \"attentive\", \"attentive\", \"attentive\", \"attentive\", \"attraction\", \"attraction\", \"availability\", \"available\", \"available\", \"available\", \"available\", \"available\", \"away\", \"away\", \"away\", \"away\", \"azadeh\", \"bakery\", \"balcony\", \"balcony\", \"bar\", \"bar\", \"bar\", \"barbara\", \"bathroom\", \"bathroom\", \"bathroom\", \"beam\", \"beauty\", \"bed\", \"bed\", \"bed\", \"bed\", \"bed\", \"bedroom\", \"bedroom\", \"big\", \"big\", \"big\", \"big\", \"big\", \"big\", \"big\", \"big\", \"biscuit\", \"bit\", \"bit\", \"bite\", \"block\", \"block\", \"boboli\", \"boiler\", \"book\", \"book\", \"book\", \"book\", \"book\", \"bother\", \"bother\", \"bottle\", \"bottle\", \"breakfast\", \"breakfast\", \"breakfast\", \"breakfast\", \"breakfast\", \"breakfast\", \"breathtake\", \"breathtaking\", \"breditz\", \"bridge\", \"bril\", \"brilliant\", \"brilliant\", \"brje\", \"brla\", \"brlappartement\", \"brown\", \"building\", \"building\", \"building\", \"building\", \"building\", \"building\", \"bus\", \"bustle\", \"busy\", \"busy\", \"cafe\", \"camilla\", \"camilla\", \"car\", \"care\", \"care\", \"caring\", \"carla\", \"carmine\", \"casa\", \"cat\", \"cathedral\", \"cathedral\", \"ceiling\", \"central\", \"central\", \"central\", \"central\", \"central\", \"central\", \"central\", \"centrale\", \"centre\", \"centre\", \"centre\", \"characteristic\", \"check\", \"check\", \"checkin\", \"checkout\", \"checkout\", \"chianti\", \"chianti\", \"chiara\", \"claudio\", \"cleanliness\", \"cleanliness\", \"cleanliness\", \"clothe\", \"coffee\", \"coffee\", \"coffee\", \"coffee\", \"cold\", \"cold\", \"come\", \"come\", \"come\", \"come\", \"come\", \"come\", \"come\", \"comfort\", \"communicate\", \"communicate\", \"communication\", \"communication\", \"complimentary\", \"conditioner\", \"conditioning\", \"conform\", \"connected\", \"convenient\", \"convenient\", \"convenient\", \"convenient\", \"convenient\", \"convenient\", \"corner\", \"corner\", \"corner\", \"correct\", \"cost\", \"countryside\", \"courtyard\", \"courtyard\", \"courtyard\", \"cream\", \"cristiana\", \"crowd\", \"daniele\", \"daniele\", \"day\", \"day\", \"day\", \"day\", \"day\", \"day\", \"day\", \"day\", \"deck\", \"decorate\", \"decorate\", \"decorate\", \"decorate\", \"decorate\", \"decoration\", \"decoration\", \"definitely\", \"definitely\", \"definitely\", \"definitely\", \"definitely\", \"delighted\", \"della\", \"describe\", \"describe\", \"describe\", \"describe\", \"describe\", \"describe\", \"deserve\", \"detail\", \"detail\", \"detail\", \"detailed\", \"detailed\", \"difficult\", \"difficult\", \"direction\", \"direction\", \"direction\", \"direction\", \"discreet\", \"dish\", \"dishwasher\", \"distance\", \"distance\", \"disturb\", \"dog\", \"dome\", \"door\", \"door\", \"door\", \"door\", \"double\", \"doubt\", \"doubt\", \"drop\", \"drop\", \"dryer\", \"duomo\", \"duomo\", \"duomo\", \"duomo\", \"ear\", \"early\", \"early\", \"early\", \"earplug\", \"easy\", \"easy\", \"easy\", \"easy\", \"eat\", \"eat\", \"eat\", \"edoardo\", \"elevator\", \"elevator\", \"elisa\", \"elisabetta\", \"enjoy\", \"enjoy\", \"enjoy\", \"enjoy\", \"enjoy\", \"enjoy\", \"equip\", \"equip\", \"equip\", \"equip\", \"equip\", \"equip\", \"exactly\", \"exactly\", \"exactly\", \"exceed\", \"expectation\", \"expectation\", \"experience\", \"experience\", \"experience\", \"experience\", \"experience\", \"explain\", \"explain\", \"explain\", \"explore\", \"explore\", \"explore\", \"explore\", \"extreme\", \"extremely\", \"extremely\", \"extremely\", \"extremely\", \"extremely\", \"extremely\", \"fabio\", \"face\", \"family\", \"family\", \"family\", \"family\", \"family\", \"family\", \"famous\", \"fantastic\", \"fantastic\", \"fantastic\", \"fantastic\", \"fantastic\", \"fast\", \"fast\", \"father\", \"federica\", \"federico\", \"federico\", \"feel\", \"feel\", \"feel\", \"feel\", \"find\", \"find\", \"find\", \"find\", \"find\", \"find\", \"find\", \"fine\", \"fine\", \"fiore\", \"firenze\", \"firenze\", \"firenze\", \"firenze\", \"firenze\", \"firenze\", \"fix\", \"flat\", \"flexible\", \"flexible\", \"floor\", \"floor\", \"floor\", \"food\", \"food\", \"food\", \"foot\", \"foot\", \"francescos\", \"fridge\", \"fridge\", \"friendly\", \"friendly\", \"friendly\", \"friendly\", \"friendly\", \"gallery\", \"garage\", \"gelato\", \"gian\", \"giancarlo\", \"giorgio\", \"giovanna\", \"giulia\", \"giulio\", \"giuseppe\", \"give\", \"give\", \"go\", \"go\", \"go\", \"go\", \"go\", \"go\", \"go\", \"gorgeous\", \"gorgeous\", \"gorgeous\", \"green\", \"greet\", \"greet\", \"greet\", \"grocery\", \"grocery\", \"guest\", \"guest\", \"guest\", \"guest\", \"guide\", \"guide\", \"guido\", \"hair\", \"hear\", \"heart\", \"heart\", \"heart\", \"heart\", \"heart\", \"heat\", \"help\", \"help\", \"help\", \"help\", \"hesitation\", \"historic\", \"historic\", \"historic\", \"historic\", \"home\", \"home\", \"hope\", \"hope\", \"hope\", \"hope\", \"hopefully\", \"hospitality\", \"hospitality\", \"hostess\", \"hostess\", \"hostess\", \"hot\", \"hot\", \"house\", \"house\", \"hustle\", \"ice\", \"ideally\", \"ideally\", \"ideally\", \"ilaria\", \"impeccable\", \"importantly\", \"incredible\", \"incredible\", \"incredible\", \"incredible\", \"independent\", \"info\", \"information\", \"internal\", \"internet\", \"irene\", \"iron\", \"isabella\", \"issue\", \"issue\", \"jacopo\", \"kind\", \"kind\", \"kind\", \"kind\", \"kindness\", \"kitchen\", \"kitchen\", \"kitchen\", \"kitchen\", \"large\", \"large\", \"large\", \"large\", \"large\", \"late\", \"late\", \"late\", \"late\", \"leave\", \"leave\", \"leave\", \"leave\", \"leave\", \"leave\", \"leonardo\", \"let\", \"let\", \"lift\", \"light\", \"light\", \"light\", \"like\", \"like\", \"like\", \"like\", \"like\", \"little\", \"little\", \"little\", \"little\", \"little\", \"little\", \"little\", \"little\", \"living\", \"living\", \"local\", \"local\", \"lock\", \"lorenzos\", \"lot\", \"lot\", \"lot\", \"lot\", \"lot\", \"lot\", \"lot\", \"lot\", \"loud\", \"love\", \"love\", \"love\", \"love\", \"lovely\", \"lovely\", \"lovely\", \"lovely\", \"luca\", \"lucas\", \"luggage\", \"luggage\", \"luigi\", \"machine\", \"magnificent\", \"main\", \"main\", \"main\", \"main\", \"main\", \"major\", \"major\", \"map\", \"marco\", \"marco\", \"maria\", \"marina\", \"mario\", \"market\", \"market\", \"market\", \"martina\", \"massimo\", \"massimo\", \"maurizio\", \"max\", \"maximum\", \"meet\", \"meet\", \"meet\", \"meet\", \"meet\", \"mercato\", \"message\", \"message\", \"meter\", \"meter\", \"mezzanine\", \"michelangelo\", \"michele\", \"microwave\", \"midnight\", \"min\", \"min\", \"min\", \"min\", \"mirko\", \"moment\", \"moment\", \"money\", \"money\", \"monument\", \"monument\", \"morning\", \"morning\", \"morning\", \"morning\", \"morning\", \"morning\", \"morning\", \"mosquito\", \"museum\", \"museum\", \"natural\", \"near\", \"near\", \"near\", \"near\", \"nearby\", \"nearby\", \"nearby\", \"nearby\", \"neighborhood\", \"neighborhood\", \"neighborhood\", \"neighborhood\", \"neighborhood\", \"neighborhood\", \"nest\", \"nicoletta\", \"night\", \"night\", \"night\", \"night\", \"night\", \"noise\", \"noisy\", \"novella\", \"oil\", \"olive\", \"open\", \"open\", \"organized\", \"outside\", \"outside\", \"outside\", \"outside\", \"outside\", \"outside\", \"palace\", \"palazzo\", \"paper\", \"park\", \"parking\", \"patrizia\", \"people\", \"people\", \"people\", \"people\", \"people\", \"people\", \"photo\", \"photo\", \"photo\", \"photo\", \"photo\", \"photo\", \"piazza\", \"pitti\", \"pizza\", \"pleasant\", \"pleasant\", \"pleasant\", \"pleasure\", \"pleasure\", \"ponte\", \"position\", \"possibility\", \"precise\", \"pressure\", \"problem\", \"problem\", \"problem\", \"process\", \"prompt\", \"provide\", \"provide\", \"provide\", \"public\", \"query\", \"question\", \"quick\", \"quickly\", \"quickly\", \"quickly\", \"quiet\", \"quiet\", \"quiet\", \"quiet\", \"quiet\", \"quiet\", \"reach\", \"reach\", \"reachable\", \"reality\", \"receive\", \"receive\", \"receive\", \"recommendation\", \"recommendation\", \"recommendation\", \"repeat\", \"reply\", \"respond\", \"response\", \"responsive\", \"responsive\", \"responsive\", \"responsive\", \"responsive\", \"responsive\", \"return\", \"return\", \"return\", \"return\", \"return\", \"riccardo\", \"right\", \"right\", \"right\", \"right\", \"right\", \"river\", \"roberta\", \"roberto\", \"roof\", \"rooftop\", \"samuel\", \"sandra\", \"santo\", \"sara\", \"say\", \"say\", \"say\", \"self\", \"self\", \"shampoo\", \"sheet\", \"shop\", \"shop\", \"shop\", \"show\", \"show\", \"show\", \"show\", \"show\", \"shower\", \"shower\", \"shutter\", \"sight\", \"sight\", \"sight\", \"sight\", \"sight\", \"signoria\", \"silvia\", \"site\", \"site\", \"site\", \"site\", \"site\", \"sleep\", \"sleeper\", \"small\", \"small\", \"small\", \"small\", \"small\", \"smooth\", \"smoothly\", \"snack\", \"snack\", \"snack\", \"soap\", \"sofa\", \"sofa\", \"solo\", \"soon\", \"soon\", \"sound\", \"sound\", \"space\", \"space\", \"space\", \"space\", \"space\", \"space\", \"spacious\", \"spacious\", \"spacious\", \"spacious\", \"spacious\", \"spacious\", \"spacious\", \"spacious\", \"special\", \"special\", \"special\", \"spectacular\", \"spend\", \"spend\", \"spend\", \"spend\", \"spend\", \"spend\", \"spirito\", \"splendid\", \"stair\", \"stair\", \"steep\", \"stop\", \"strategic\", \"street\", \"street\", \"stroll\", \"strong\", \"structure\", \"stunning\", \"stylish\", \"stylish\", \"stylish\", \"suggestion\", \"suggestion\", \"suggestion\", \"suitcase\", \"suitcase\", \"summer\", \"summer\", \"sunset\", \"super\", \"super\", \"super\", \"super\", \"super\", \"supermarket\", \"supermarket\", \"supermarket\", \"supermarket\", \"supermarket\", \"surely\", \"system\", \"tall\", \"taste\", \"taste\", \"terrace\", \"thing\", \"thing\", \"thing\", \"thing\", \"thing\", \"thing\", \"thoroughly\", \"timely\", \"tip\", \"tiziana\", \"toilet\", \"tommaso\", \"tour\", \"tour\", \"tour\", \"tourist\", \"tourist\", \"tourist\", \"touristy\", \"towel\", \"train\", \"train\", \"train\", \"train\", \"train\", \"train\", \"tram\", \"transport\", \"tuck\", \"uffizi\", \"use\", \"use\", \"use\", \"use\", \"use\", \"useful\", \"useful\", \"useful\", \"valeria\", \"valuable\", \"value\", \"value\", \"value\", \"vecchio\", \"view\", \"visit\", \"visit\", \"visit\", \"visit\", \"visit\", \"visit\", \"wait\", \"wait\", \"wait\", \"wait\", \"walking\", \"walking\", \"want\", \"want\", \"want\", \"want\", \"want\", \"want\", \"want\", \"want\", \"warm\", \"warm\", \"warm\", \"warm\", \"wash\", \"washing\", \"water\", \"water\", \"welcome\", \"welcome\", \"welcome\", \"wifi\", \"wifi\", \"wifi\", \"window\", \"window\", \"wine\", \"wine\", \"wine\", \"wine\", \"winter\", \"wonderful\", \"wonderful\", \"wonderful\", \"wonderful\", \"wonderful\", \"work\", \"work\", \"work\", \"worth\", \"worth\", \"ztl\"]}, \"R\": 30, \"lambda.step\": 0.01, \"plot.opts\": {\"xlab\": \"PC1\", \"ylab\": \"PC2\"}, \"topic.order\": [1, 2, 3, 4, 5, 6, 7, 8]};\n",
       "\n",
       "function LDAvis_load_lib(url, callback){\n",
       "  var s = document.createElement('script');\n",
       "  s.src = url;\n",
       "  s.async = true;\n",
       "  s.onreadystatechange = s.onload = callback;\n",
       "  s.onerror = function(){console.warn(\"failed to load library \" + url);};\n",
       "  document.getElementsByTagName(\"head\")[0].appendChild(s);\n",
       "}\n",
       "\n",
       "if(typeof(LDAvis) !== \"undefined\"){\n",
       "   // already loaded: just create the visualization\n",
       "   !function(LDAvis){\n",
       "       new LDAvis(\"#\" + \"ldavis_el7713859980289449921707526\", ldavis_el7713859980289449921707526_data);\n",
       "   }(LDAvis);\n",
       "}else if(typeof define === \"function\" && define.amd){\n",
       "   // require.js is available: use it to load d3/LDAvis\n",
       "   require.config({paths: {d3: \"https://d3js.org/d3.v5\"}});\n",
       "   require([\"d3\"], function(d3){\n",
       "      window.d3 = d3;\n",
       "      LDAvis_load_lib(\"https://cdn.jsdelivr.net/gh/bmabey/pyLDAvis@3.3.1/pyLDAvis/js/ldavis.v3.0.0.js\", function(){\n",
       "        new LDAvis(\"#\" + \"ldavis_el7713859980289449921707526\", ldavis_el7713859980289449921707526_data);\n",
       "      });\n",
       "    });\n",
       "}else{\n",
       "    // require.js not available: dynamically load d3 & LDAvis\n",
       "    LDAvis_load_lib(\"https://d3js.org/d3.v5.js\", function(){\n",
       "         LDAvis_load_lib(\"https://cdn.jsdelivr.net/gh/bmabey/pyLDAvis@3.3.1/pyLDAvis/js/ldavis.v3.0.0.js\", function(){\n",
       "                 new LDAvis(\"#\" + \"ldavis_el7713859980289449921707526\", ldavis_el7713859980289449921707526_data);\n",
       "            })\n",
       "         });\n",
       "}\n",
       "</script>"
      ],
      "text/plain": [
       "PreparedData(topic_coordinates=              x         y  topics  cluster       Freq\n",
       "topic                                                \n",
       "0     -0.116582  0.287955       1        1   8.874749\n",
       "1     -0.033068  0.135506       2        1  15.833814\n",
       "2     -0.165427 -0.158472       3        1  11.153423\n",
       "3      0.048799 -0.126202       4        1  10.432188\n",
       "4     -0.131129 -0.000426       5        1  12.780142\n",
       "5      0.198828 -0.103341       6        1  17.012538\n",
       "6      0.312008  0.081124       7        1  10.395133\n",
       "7     -0.113429 -0.116145       8        1  13.518010, topic_info=          Term         Freq        Total Category  logprob  loglift\n",
       "90        view  4253.000000  4253.000000  Default  30.0000  30.0000\n",
       "144       give  4642.000000  4642.000000  Default  29.0000  29.0000\n",
       "78       house  5799.000000  5799.000000  Default  28.0000  28.0000\n",
       "562       flat  3305.000000  3305.000000  Default  27.0000  27.0000\n",
       "130      check  4089.000000  4089.000000  Default  26.0000  26.0000\n",
       "..         ...          ...          ...      ...      ...      ...\n",
       "147      super   746.734633  3615.999702   Topic8  -5.0320   0.4237\n",
       "1    available   670.074609  2407.595548   Topic8  -5.1403   0.7222\n",
       "208      photo   631.584319  1806.624424   Topic8  -5.1995   0.9502\n",
       "271       help   641.213539  3124.777117   Topic8  -5.1843   0.4174\n",
       "252      equip   629.968811  2723.503882   Topic8  -5.2020   0.5371\n",
       "\n",
       "[531 rows x 6 columns], token_table=      Topic      Freq        Term\n",
       "term                             \n",
       "330       1  0.462410  absolutely\n",
       "330       2  0.086229  absolutely\n",
       "330       5  0.201202  absolutely\n",
       "330       8  0.249611  absolutely\n",
       "526       1  0.181309  accessible\n",
       "...     ...       ...         ...\n",
       "308       6  0.765538        work\n",
       "308       7  0.111092        work\n",
       "63        1  0.719795       worth\n",
       "63        6  0.279481       worth\n",
       "1674      7  0.992044         ztl\n",
       "\n",
       "[956 rows x 3 columns], R=30, lambda_step=0.01, plot_opts={'xlab': 'PC1', 'ylab': 'PC2'}, topic_order=[1, 2, 3, 4, 5, 6, 7, 8])"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# To plot inside Jupyter notebook\n",
    "pyLDAvis.enable_notebook()\n",
    "\n",
    "LDAvis_prepared = pyLDAvis.gensim_models.prepare(lda_model_final, corpus, id2word, sort_topics=False)\n",
    "\n",
    "# Save pyLDA plot as html file\n",
    "pyLDAvis.save_html(LDAvis_prepared, './ldavis_prepared_'+ str(num_topics) +'.html')\n",
    "LDAvis_prepared"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b9031aa-d5b6-4672-a9b0-16fc564cbe6e",
   "metadata": {},
   "source": [
    "Each bubble on the left-hand side of the plot represents an individual topic. The **larger** the **bubble**, the **more important** is the **topic**.<br>\n",
    "The **distance between the topics** is an approximation of **semantic relationship** between them. Indeed, the topics which share common words are overlapping (closer in distance).\n",
    "Ideally, a good topic model will have fairly big, non-overlapping bubbles scattered throughout the chart instead of being clustered in one.<br>\n",
    "\n",
    "On the right-hand side, the **horizontal bar chart** represents the **words distribution a topic decreasing order of relevance**. The **red bar** indicates the **frequency** of the word **within the selected topic**, while the **blue bar** indicates the **frequency within the corpus** of reviews. <br>\n",
    "\n",
    "**Adjusting λ** (lambda) <br>\n",
    "If λ=1, the relevance is given purely by the probability of the word to that topic. <br>\n",
    "If λ=0, the relevance is given by the probability of a word appearing in a particular topic divided by the probability of the word appearing generally — thus, highly frequent words (such as ‘like’, ‘good’) will be downgraded heavily in relevance when we have a lower λ value.\n",
    "\n",
    "\n",
    "**Some insights**:<br>\n",
    "\n",
    "- The most important topic is **Accomodation features**\n",
    "- The least important topic is **Good feelings**\n",
    "- There is a bit of overalap between topic 3 **Tips** and topic 8 **Good experience**\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "b8617ae0-bc4e-4a2c-968a-474603f8cbe4",
   "metadata": {},
   "outputs": [],
   "source": [
    "#df_reviews_sample_sentiment_bitrigrams.to_pickle('./data/df_reviews_sample_topics.pkl')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b7964205-7c5f-47a0-b910-f490533d4197",
   "metadata": {},
   "source": [
    "[Back to TOC](#Table-of-Contents:)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ec2681c5-a904-468e-9ba3-03e569c9e056",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}