DBSCAN_Clustering_with_hyperparameter_tuning.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reading CSV file\n",
    "\n",
    "df= pd.read_csv('Customers.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>CustomerID</th>\n",
       "      <th>Genre</th>\n",
       "      <th>Age</th>\n",
       "      <th>Annual Income (k$)</th>\n",
       "      <th>Spending Score (1-100)</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Male</td>\n",
       "      <td>19</td>\n",
       "      <td>15</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>Male</td>\n",
       "      <td>21</td>\n",
       "      <td>15</td>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Female</td>\n",
       "      <td>20</td>\n",
       "      <td>16</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>Female</td>\n",
       "      <td>23</td>\n",
       "      <td>16</td>\n",
       "      <td>77</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>Female</td>\n",
       "      <td>31</td>\n",
       "      <td>17</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>195</th>\n",
       "      <td>196</td>\n",
       "      <td>Female</td>\n",
       "      <td>35</td>\n",
       "      <td>120</td>\n",
       "      <td>79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>196</th>\n",
       "      <td>197</td>\n",
       "      <td>Female</td>\n",
       "      <td>45</td>\n",
       "      <td>126</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>197</th>\n",
       "      <td>198</td>\n",
       "      <td>Male</td>\n",
       "      <td>32</td>\n",
       "      <td>126</td>\n",
       "      <td>74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>198</th>\n",
       "      <td>199</td>\n",
       "      <td>Male</td>\n",
       "      <td>32</td>\n",
       "      <td>137</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>199</th>\n",
       "      <td>200</td>\n",
       "      <td>Male</td>\n",
       "      <td>30</td>\n",
       "      <td>137</td>\n",
       "      <td>83</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>200 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     CustomerID   Genre  Age  Annual Income (k$)  Spending Score (1-100)\n",
       "0             1    Male   19                  15                      39\n",
       "1             2    Male   21                  15                      81\n",
       "2             3  Female   20                  16                       6\n",
       "3             4  Female   23                  16                      77\n",
       "4             5  Female   31                  17                      40\n",
       "..          ...     ...  ...                 ...                     ...\n",
       "195         196  Female   35                 120                      79\n",
       "196         197  Female   45                 126                      28\n",
       "197         198    Male   32                 126                      74\n",
       "198         199    Male   32                 137                      18\n",
       "199         200    Male   30                 137                      83\n",
       "\n",
       "[200 rows x 5 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "X=df.iloc[:,[3,4]].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 15,  39],\n",
       "       [ 15,  81],\n",
       "       [ 16,   6],\n",
       "       [ 16,  77],\n",
       "       [ 17,  40],\n",
       "       [ 17,  76],\n",
       "       [ 18,   6],\n",
       "       [ 18,  94],\n",
       "       [ 19,   3],\n",
       "       [ 19,  72],\n",
       "       [ 19,  14],\n",
       "       [ 19,  99],\n",
       "       [ 20,  15],\n",
       "       [ 20,  77],\n",
       "       [ 20,  13],\n",
       "       [ 20,  79],\n",
       "       [ 21,  35],\n",
       "       [ 21,  66],\n",
       "       [ 23,  29],\n",
       "       [ 23,  98],\n",
       "       [ 24,  35],\n",
       "       [ 24,  73],\n",
       "       [ 25,   5],\n",
       "       [ 25,  73],\n",
       "       [ 28,  14],\n",
       "       [ 28,  82],\n",
       "       [ 28,  32],\n",
       "       [ 28,  61],\n",
       "       [ 29,  31],\n",
       "       [ 29,  87],\n",
       "       [ 30,   4],\n",
       "       [ 30,  73],\n",
       "       [ 33,   4],\n",
       "       [ 33,  92],\n",
       "       [ 33,  14],\n",
       "       [ 33,  81],\n",
       "       [ 34,  17],\n",
       "       [ 34,  73],\n",
       "       [ 37,  26],\n",
       "       [ 37,  75],\n",
       "       [ 38,  35],\n",
       "       [ 38,  92],\n",
       "       [ 39,  36],\n",
       "       [ 39,  61],\n",
       "       [ 39,  28],\n",
       "       [ 39,  65],\n",
       "       [ 40,  55],\n",
       "       [ 40,  47],\n",
       "       [ 40,  42],\n",
       "       [ 40,  42],\n",
       "       [ 42,  52],\n",
       "       [ 42,  60],\n",
       "       [ 43,  54],\n",
       "       [ 43,  60],\n",
       "       [ 43,  45],\n",
       "       [ 43,  41],\n",
       "       [ 44,  50],\n",
       "       [ 44,  46],\n",
       "       [ 46,  51],\n",
       "       [ 46,  46],\n",
       "       [ 46,  56],\n",
       "       [ 46,  55],\n",
       "       [ 47,  52],\n",
       "       [ 47,  59],\n",
       "       [ 48,  51],\n",
       "       [ 48,  59],\n",
       "       [ 48,  50],\n",
       "       [ 48,  48],\n",
       "       [ 48,  59],\n",
       "       [ 48,  47],\n",
       "       [ 49,  55],\n",
       "       [ 49,  42],\n",
       "       [ 50,  49],\n",
       "       [ 50,  56],\n",
       "       [ 54,  47],\n",
       "       [ 54,  54],\n",
       "       [ 54,  53],\n",
       "       [ 54,  48],\n",
       "       [ 54,  52],\n",
       "       [ 54,  42],\n",
       "       [ 54,  51],\n",
       "       [ 54,  55],\n",
       "       [ 54,  41],\n",
       "       [ 54,  44],\n",
       "       [ 54,  57],\n",
       "       [ 54,  46],\n",
       "       [ 57,  58],\n",
       "       [ 57,  55],\n",
       "       [ 58,  60],\n",
       "       [ 58,  46],\n",
       "       [ 59,  55],\n",
       "       [ 59,  41],\n",
       "       [ 60,  49],\n",
       "       [ 60,  40],\n",
       "       [ 60,  42],\n",
       "       [ 60,  52],\n",
       "       [ 60,  47],\n",
       "       [ 60,  50],\n",
       "       [ 61,  42],\n",
       "       [ 61,  49],\n",
       "       [ 62,  41],\n",
       "       [ 62,  48],\n",
       "       [ 62,  59],\n",
       "       [ 62,  55],\n",
       "       [ 62,  56],\n",
       "       [ 62,  42],\n",
       "       [ 63,  50],\n",
       "       [ 63,  46],\n",
       "       [ 63,  43],\n",
       "       [ 63,  48],\n",
       "       [ 63,  52],\n",
       "       [ 63,  54],\n",
       "       [ 64,  42],\n",
       "       [ 64,  46],\n",
       "       [ 65,  48],\n",
       "       [ 65,  50],\n",
       "       [ 65,  43],\n",
       "       [ 65,  59],\n",
       "       [ 67,  43],\n",
       "       [ 67,  57],\n",
       "       [ 67,  56],\n",
       "       [ 67,  40],\n",
       "       [ 69,  58],\n",
       "       [ 69,  91],\n",
       "       [ 70,  29],\n",
       "       [ 70,  77],\n",
       "       [ 71,  35],\n",
       "       [ 71,  95],\n",
       "       [ 71,  11],\n",
       "       [ 71,  75],\n",
       "       [ 71,   9],\n",
       "       [ 71,  75],\n",
       "       [ 72,  34],\n",
       "       [ 72,  71],\n",
       "       [ 73,   5],\n",
       "       [ 73,  88],\n",
       "       [ 73,   7],\n",
       "       [ 73,  73],\n",
       "       [ 74,  10],\n",
       "       [ 74,  72],\n",
       "       [ 75,   5],\n",
       "       [ 75,  93],\n",
       "       [ 76,  40],\n",
       "       [ 76,  87],\n",
       "       [ 77,  12],\n",
       "       [ 77,  97],\n",
       "       [ 77,  36],\n",
       "       [ 77,  74],\n",
       "       [ 78,  22],\n",
       "       [ 78,  90],\n",
       "       [ 78,  17],\n",
       "       [ 78,  88],\n",
       "       [ 78,  20],\n",
       "       [ 78,  76],\n",
       "       [ 78,  16],\n",
       "       [ 78,  89],\n",
       "       [ 78,   1],\n",
       "       [ 78,  78],\n",
       "       [ 78,   1],\n",
       "       [ 78,  73],\n",
       "       [ 79,  35],\n",
       "       [ 79,  83],\n",
       "       [ 81,   5],\n",
       "       [ 81,  93],\n",
       "       [ 85,  26],\n",
       "       [ 85,  75],\n",
       "       [ 86,  20],\n",
       "       [ 86,  95],\n",
       "       [ 87,  27],\n",
       "       [ 87,  63],\n",
       "       [ 87,  13],\n",
       "       [ 87,  75],\n",
       "       [ 87,  10],\n",
       "       [ 87,  92],\n",
       "       [ 88,  13],\n",
       "       [ 88,  86],\n",
       "       [ 88,  15],\n",
       "       [ 88,  69],\n",
       "       [ 93,  14],\n",
       "       [ 93,  90],\n",
       "       [ 97,  32],\n",
       "       [ 97,  86],\n",
       "       [ 98,  15],\n",
       "       [ 98,  88],\n",
       "       [ 99,  39],\n",
       "       [ 99,  97],\n",
       "       [101,  24],\n",
       "       [101,  68],\n",
       "       [103,  17],\n",
       "       [103,  85],\n",
       "       [103,  23],\n",
       "       [103,  69],\n",
       "       [113,   8],\n",
       "       [113,  91],\n",
       "       [120,  16],\n",
       "       [120,  79],\n",
       "       [126,  28],\n",
       "       [126,  74],\n",
       "       [137,  18],\n",
       "       [137,  83]], dtype=int64)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Hyper parameter tuning "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "DBSCAN is a density based apporach \n",
    "\n",
    "The hyper parameters in DBSCAN are epsilon and min_sample\n",
    "\n",
    "min_sample we choose it basis on the data\n",
    "\n",
    "thumb rule for min_sample is ===> minpoints>= dimension of data set+1\n",
    "\n",
    "we choose larger minpoints if our dataset have more noisy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Epsilon is most important hyperparamter \n",
    "\n",
    "we calculate this with help of elbow method \n",
    "\n",
    "we plot graph between distance between points and epsilon \n",
    "\n",
    "where the plot suddenly rise we take that point as epsilon \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "# we use nearestneighbors for calculating distance between points\n",
    "from sklearn.neighbors import NearestNeighbors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# calculating distances\n",
    "neigh=NearestNeighbors(n_neighbors=2)\n",
    "distance=neigh.fit(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "# indices and distance values\n",
    "distances,indices=distance.kneighbors(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Now sorting the distance increasing order\n",
    "\n",
    "sorting_distances=np.sort(distances,axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sorted distances\n",
    "sorted_distances=sort_distances[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot between distance vs epsilon\n",
    "\n",
    "plt.plot(sorted_distances)\n",
    "plt.xlabel('Distance')\n",
    "plt.ylabel('Epsilon')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If we observe the grpah at epsilon is equal to 9 graph continuous increases so we take epsilon equal to 9"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## DBSCAN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.cluster import DBSCAN\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {},
   "outputs": [],
   "source": [
    "# intializing DBSCAN\n",
    "clustering_model=DBSCAN(eps=9,min_samples=4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DBSCAN(eps=9, min_samples=4)"
      ]
     },
     "execution_count": 181,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# fit the model to X\n",
    "clustering_model.fit(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "metadata": {},
   "outputs": [],
   "source": [
    "# predicted labels by DBSCAN\n",
    "predicted_labels=clustering_model.labels_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 183,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'DBSCAN')"
      ]
     },
     "execution_count": 183,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(X[:,0], X[:,1],c=predicted_labels, cmap='Paired')\n",
    "plt.xlabel('Annual Income')\n",
    "plt.ylabel('Spending Score')\n",
    "plt.title(\"DBSCAN\")\n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import metrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 185,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4259680122384905"
      ]
     },
     "execution_count": 185,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metrics.silhouette_score(X, predicted_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}