Candidate_Hire_Prediction_DecisionTree.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Candidate Hiring Prediction using Decision Tree."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this project, we build a Decision Tree Classifier to predict whether a candinate applying for a job at a company would get hired or not. we build two models, one with criterion gini index and another one with criterion entropy.We implement Decision Tree Classification with Python and Scikit-Learn. We have used the PastHires dataset for this project, downloaded from the UCI Machine Learning Repository website.\n",
    "\n",
    "Dataset have 7 columns, Years Experience, Employed?\tPrevious employers,\tLevel of Education,\tTop-tier school, Interned, and\tHired.\n",
    "\n",
    "Model will predict whether the candidate would get hired by company based on the attributes of the candidate who gets hired or does not get hired from the past.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Introduction to Decision Tree algorithm\n",
    "\n",
    "\n",
    "A Decision Tree algorithm is one of the most popular machine learning algorithms. It uses a tree like structure and their possible combinations to solve a particular problem. It belongs to the class of supervised learning algorithms where it can be used for both classification and regression purposes. \n",
    "\n",
    "\n",
    "A decision tree is a structure that includes a root node, branches, and leaf nodes. Each internal node denotes a test on an attribute, each branch denotes the outcome of a test, and each leaf node holds a class label. The topmost node in the tree is the root node. \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>Years Experience</th>\n",
       "      <th>Employed?</th>\n",
       "      <th>Previous employers</th>\n",
       "      <th>Level of Education</th>\n",
       "      <th>Top-tier school</th>\n",
       "      <th>Interned</th>\n",
       "      <th>Hired</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>Y</td>\n",
       "      <td>4</td>\n",
       "      <td>BS</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>0</td>\n",
       "      <td>BS</td>\n",
       "      <td>Y</td>\n",
       "      <td>Y</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>N</td>\n",
       "      <td>6</td>\n",
       "      <td>BS</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>Y</td>\n",
       "      <td>1</td>\n",
       "      <td>MS</td>\n",
       "      <td>Y</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>20</td>\n",
       "      <td>N</td>\n",
       "      <td>2</td>\n",
       "      <td>PhD</td>\n",
       "      <td>Y</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>0</td>\n",
       "      <td>PhD</td>\n",
       "      <td>Y</td>\n",
       "      <td>Y</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>Y</td>\n",
       "      <td>2</td>\n",
       "      <td>MS</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "      <td>N</td>\n",
       "      <td>1</td>\n",
       "      <td>BS</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>15</td>\n",
       "      <td>Y</td>\n",
       "      <td>5</td>\n",
       "      <td>BS</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>N</td>\n",
       "      <td>0</td>\n",
       "      <td>BS</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Years Experience Employed?  Previous employers Level of Education  \\\n",
       "0                10         Y                   4                 BS   \n",
       "1                 0         N                   0                 BS   \n",
       "2                 7         N                   6                 BS   \n",
       "3                 2         Y                   1                 MS   \n",
       "4                20         N                   2                PhD   \n",
       "5                 0         N                   0                PhD   \n",
       "6                 5         Y                   2                 MS   \n",
       "7                 3         N                   1                 BS   \n",
       "8                15         Y                   5                 BS   \n",
       "9                 0         N                   0                 BS   \n",
       "\n",
       "  Top-tier school Interned Hired  \n",
       "0               N        N     Y  \n",
       "1               Y        Y     Y  \n",
       "2               N        N     N  \n",
       "3               Y        N     Y  \n",
       "4               Y        N     N  \n",
       "5               Y        Y     Y  \n",
       "6               N        Y     Y  \n",
       "7               N        Y     Y  \n",
       "8               N        N     Y  \n",
       "9               N        N     N  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn import tree\n",
    "\n",
    " \n",
    "df = pd.read_csv(\"C:\\\\Users\\\\LENOVO\\\\Desktop\\\\ML\\\\DMBI\\\\MY_DT\\\\PastHires.csv\")\n",
    "\n",
    "df.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "scikit-learn needs everything to be numerical for decision trees to work. So, we mapped Y,N to 1,0 and levels of education to some scale of 0-2."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>Years Experience</th>\n",
       "      <th>Employed?</th>\n",
       "      <th>Previous employers</th>\n",
       "      <th>Level of Education</th>\n",
       "      <th>Top-tier school</th>\n",
       "      <th>Interned</th>\n",
       "      <th>Hired</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>20</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Years Experience  Employed?  Previous employers  Level of Education  \\\n",
       "0                10          1                   4                   0   \n",
       "1                 0          0                   0                   0   \n",
       "2                 7          0                   6                   0   \n",
       "3                 2          1                   1                   1   \n",
       "4                20          0                   2                   2   \n",
       "\n",
       "   Top-tier school  Interned  Hired  \n",
       "0                0         0      1  \n",
       "1                1         1      1  \n",
       "2                0         0      0  \n",
       "3                1         0      1  \n",
       "4                1         0      0  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "d = {'Y': 1, 'N': 0}\n",
    "df['Hired'] = df['Hired'].map(d)\n",
    "df['Employed?'] = df['Employed?'].map(d)\n",
    "df['Top-tier school'] = df['Top-tier school'].map(d)\n",
    "df['Interned'] = df['Interned'].map(d)\n",
    "d = {'BS': 0, 'MS': 1, 'PhD': 2}\n",
    "df['Level of Education'] = df['Level of Education'].map(d)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Next we  separate the features from the target column that we're trying to bulid a decision tree for."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Years Experience',\n",
       " 'Employed?',\n",
       " 'Previous employers',\n",
       " 'Level of Education',\n",
       " 'Top-tier school',\n",
       " 'Interned']"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features = list(df.columns[:6])\n",
    "features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = df[\"Hired\"]\n",
    "X = df[features]\n",
    "clf = tree.DecisionTreeClassifier()\n",
    "clf = clf.fit(X,y)   #fitting"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Constructing Decision Tree."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.display import Image  \n",
    "from sklearn.externals.six import StringIO  \n",
    "import pydotplus\n",
    "\n",
    "dot_data = StringIO()  \n",
    "tree.export_graphviz(clf, out_file=dot_data,  \n",
    "                         feature_names=features)  \n",
    "graph = pydotplus.graph_from_dot_data(dot_data.getvalue())  \n",
    "Image(graph.create_png())  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Split data into separate training and test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.20, random_state = 50)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((27, 6), (7, 6))"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check the shape of X_train and X_test\n",
    "\n",
    "X_train.shape, X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Years Experience      int64\n",
       "Employed?             int64\n",
       "Previous employers    int64\n",
       "Level of Education    int64\n",
       "Top-tier school       int64\n",
       "Interned              int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check data types in X_train\n",
    "\n",
    "X_train.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Encode categorical variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import category encoders\n",
    "\n",
    "import category_encoders as ce"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "encoder = ce.OrdinalEncoder(cols=['Years Experience', 'Employed?', 'Previous employers', 'Level of Education', 'Top-tier school', 'Interned'])\n",
    "\n",
    "\n",
    "X_train = encoder.fit_transform(X_train)\n",
    "\n",
    "X_test = encoder.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(class_weight=None, criterion='gini', max_depth=3,\n",
       "                       max_features=None, max_leaf_nodes=None,\n",
       "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                       min_samples_leaf=1, min_samples_split=2,\n",
       "                       min_weight_fraction_leaf=0.0, presort=False,\n",
       "                       random_state=0, splitter='best')"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# import DecisionTreeClassifier\n",
    "\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "clf_gini = DecisionTreeClassifier(criterion='gini', max_depth=3, random_state=0)\n",
    "\n",
    "\n",
    "# fit the model\n",
    "clf_gini.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predict the Test set results with criterion gini index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_gini = clf_gini.predict(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Check accuracy score with criterion gini index."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model accuracy score with criterion gini index: 0.7143\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "print('Model accuracy score with criterion gini index: {0:0.4f}'. format(accuracy_score(y_test, y_pred_gini)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training-set accuracy score: 0.9259\n"
     ]
    }
   ],
   "source": [
    "y_pred_train_gini = clf_gini.predict(X_train)\n",
    "\n",
    "y_pred_train_gini\n",
    "print('Training-set accuracy score: {0:0.4f}'. format(accuracy_score(y_train, y_pred_train_gini)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Decision Tree Classifier with criterion entropy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(class_weight=None, criterion='entropy', max_depth=3,\n",
       "                       max_features=None, max_leaf_nodes=None,\n",
       "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                       min_samples_leaf=1, min_samples_split=2,\n",
       "                       min_weight_fraction_leaf=0.0, presort=False,\n",
       "                       random_state=0, splitter='best')"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# instantiate the DecisionTreeClassifier model with criterion entropy\n",
    "\n",
    "clf_en = DecisionTreeClassifier(criterion='entropy', max_depth=3, random_state=0)\n",
    "\n",
    "\n",
    "# fit the model\n",
    "clf_en.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model accuracy score with criterion entropy: 0.7143\n"
     ]
    }
   ],
   "source": [
    "y_pred_en = clf_en.predict(X_test)\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "print('Model accuracy score with criterion entropy: {0:0.4f}'. format(accuracy_score(y_test, y_pred_en)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training-set accuracy score: 0.9259\n"
     ]
    }
   ],
   "source": [
    "y_pred_train_en = clf_en.predict(X_train)\n",
    "\n",
    "y_pred_train_en\n",
    "print('Training-set accuracy score: {0:0.4f}'. format(accuracy_score(y_train, y_pred_train_en)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Confusion matrix\n",
    "A confusion matrix is a tool for summarizing the performance of a classification algorithm. A confusion matrix will give us a clear picture of classification model performance and the types of errors produced by the model. It gives us a summary of correct and incorrect predictions broken down by each category. The summary is represented in a tabular form."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Confusion matrix\n",
      "\n",
      " [[2 0]\n",
      " [2 3]]\n"
     ]
    }
   ],
   "source": [
    "# Print the Confusion Matrix and slice it into four pieces\n",
    "\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "cm = confusion_matrix(y_test, y_pred_en)\n",
    "\n",
    "print('Confusion matrix\\n\\n', cm)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Classification Report\n",
    "\n",
    "Classification report is another way to evaluate the classification model performance. It displays the precision, recall, f1 and support scores for the model. I have described these terms in later.\n",
    "\n",
    "We can print a classification report as follows:-"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.50      1.00      0.67         2\n",
      "           1       1.00      0.60      0.75         5\n",
      "\n",
      "    accuracy                           0.71         7\n",
      "   macro avg       0.75      0.80      0.71         7\n",
      "weighted avg       0.86      0.71      0.73         7\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import classification_report\n",
    "\n",
    "print(classification_report(y_test, y_pred_en))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Ensemble learning: using a random forest"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## We use a random forest of 50 decision trees to predict employment of specific candidate profiles."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of year  Experience: 12\n",
      "Employed?: 0\n",
      "Enter Number of Previous employers: 2\n",
      "Level of Education: 1\n",
      " From Top-tier school?: 1\n",
      "Interned in this company?: 1\n",
      "['12', '0', '2', '1', '1', '1']\n",
      "Candidate can get Hired.\n"
     ]
    }
   ],
   "source": [
    "pList = [] \n",
    "maxLengthList = 6\n",
    "while len(pList) < maxLengthList:\n",
    "    exp = input(\"Number of year  Experience: \")\n",
    "    pList.append(exp)\n",
    "    emp = input(\"Employed?: \") #binary input 0 for No, 1 for Yes.\n",
    "    pList.append(emp)\n",
    "    pemp = input(\"Enter Number of Previous employers: \")\n",
    "    pList.append(pemp)\n",
    "    edu = input(\"Level of Education: \") # 0 for bachelors, 1 for Masters, 2 for PhDs \n",
    "    pList.append(edu)\n",
    "    top = input(\" From Top-tier school?: \") #binary input 0 for No, 1 for Yes. \n",
    "    pList.append(top)\n",
    "    intern = input(\"Interned in this company?: \") #binary input 0 for No, 1 for Yes.\n",
    "    pList.append(intern)\n",
    "print(pList)\n",
    "\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "clf = RandomForestClassifier(n_estimators=10)\n",
    "clf = clf.fit(X, y)\n",
    "\n",
    "\n",
    "if (clf.predict([pList])) == 0:\n",
    "    print(\"Candidate can not get Hired.\")\n",
    "else:\n",
    "    print(\"Candidate can get Hired.\")"
   ]
  }
 ],
 "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}