RandomForest_Classifier.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Collection of Data**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "legitimate_urls = pd.read_csv(\"C:extracted_csv_files/legitimate-urls.csv\")\n",
    "phishing_urls = pd.read_csv(\"C:extracted_csv_files/phishing-urls.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>Domain</th>\n",
       "      <th>Having_@_symbol</th>\n",
       "      <th>Having_IP</th>\n",
       "      <th>Path</th>\n",
       "      <th>Prefix_suffix_separation</th>\n",
       "      <th>Protocol</th>\n",
       "      <th>Redirection_//_symbol</th>\n",
       "      <th>Sub_domains</th>\n",
       "      <th>URL_Length</th>\n",
       "      <th>age_domain</th>\n",
       "      <th>dns_record</th>\n",
       "      <th>domain_registration_length</th>\n",
       "      <th>http_tokens</th>\n",
       "      <th>label</th>\n",
       "      <th>statistical_report</th>\n",
       "      <th>tiny_url</th>\n",
       "      <th>web_traffic</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>asesoresvelfit.com</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>/media/datacredito.co/</td>\n",
       "      <td>0</td>\n",
       "      <td>http</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>caixa.com.br.fgtsagendesaqueconta.com</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>/consulta8523211/principal.php</td>\n",
       "      <td>0</td>\n",
       "      <td>http</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>hissoulreason.com</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>/js/homepage/home/</td>\n",
       "      <td>0</td>\n",
       "      <td>http</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>unauthorizd.newebpage.com</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>/webapps/66fbf/</td>\n",
       "      <td>0</td>\n",
       "      <td>http</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>133.130.103.10</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>/23/</td>\n",
       "      <td>0</td>\n",
       "      <td>http</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>dj00.co.vu</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>/css/</td>\n",
       "      <td>0</td>\n",
       "      <td>http</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>133.130.103.10</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>/21/logar/</td>\n",
       "      <td>0</td>\n",
       "      <td>http</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>httpssicredi.esy.es</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>/servico/sicredi/validarclientes/mobi/index.php</td>\n",
       "      <td>0</td>\n",
       "      <td>http</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</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",
       "      <th>8</th>\n",
       "      <td>gamesaty.ga</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>/wp-content///yh/en/</td>\n",
       "      <td>0</td>\n",
       "      <td>http</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>luxuryupgradepro.com</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>/ymailNew/ymailNew/</td>\n",
       "      <td>0</td>\n",
       "      <td>http</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  Domain  Having_@_symbol  Having_IP  \\\n",
       "0                     asesoresvelfit.com                0          0   \n",
       "1  caixa.com.br.fgtsagendesaqueconta.com                0          0   \n",
       "2                      hissoulreason.com                0          0   \n",
       "3              unauthorizd.newebpage.com                0          0   \n",
       "4                         133.130.103.10                0          1   \n",
       "5                             dj00.co.vu                1          0   \n",
       "6                         133.130.103.10                0          1   \n",
       "7                    httpssicredi.esy.es                0          0   \n",
       "8                            gamesaty.ga                0          0   \n",
       "9                   luxuryupgradepro.com                0          0   \n",
       "\n",
       "                                              Path  Prefix_suffix_separation  \\\n",
       "0                           /media/datacredito.co/                         0   \n",
       "1                   /consulta8523211/principal.php                         0   \n",
       "2                               /js/homepage/home/                         0   \n",
       "3                                  /webapps/66fbf/                         0   \n",
       "4                                             /23/                         0   \n",
       "5                                            /css/                         0   \n",
       "6                                       /21/logar/                         0   \n",
       "7  /servico/sicredi/validarclientes/mobi/index.php                         0   \n",
       "8                             /wp-content///yh/en/                         0   \n",
       "9                              /ymailNew/ymailNew/                         0   \n",
       "\n",
       "  Protocol  Redirection_//_symbol  Sub_domains  URL_Length  age_domain  \\\n",
       "0     http                      0            0           0           0   \n",
       "1     http                      0            1           1           0   \n",
       "2     http                      0            0           0           0   \n",
       "3     http                      0            0           0           0   \n",
       "4     http                      0            2           0           1   \n",
       "5     http                      0            0           2           1   \n",
       "6     http                      0            2           0           1   \n",
       "7     http                      0            2           2           1   \n",
       "8     http                      1            0           2           1   \n",
       "9     http                      0            0           0           0   \n",
       "\n",
       "   dns_record  domain_registration_length  http_tokens  label  \\\n",
       "0           0                           1            0      1   \n",
       "1           0                           1            0      1   \n",
       "2           0                           1            0      1   \n",
       "3           0                           1            0      1   \n",
       "4           0                           1            0      1   \n",
       "5           1                           1            0      1   \n",
       "6           0                           1            0      1   \n",
       "7           1                           1            1      1   \n",
       "8           0                           1            0      1   \n",
       "9           0                           1            0      1   \n",
       "\n",
       "   statistical_report  tiny_url  web_traffic  \n",
       "0                   0         1            1  \n",
       "1                   1         0            1  \n",
       "2                   0         0            1  \n",
       "3                   1         0            1  \n",
       "4                   0         0            1  \n",
       "5                   1         0            0  \n",
       "6                   0         0            1  \n",
       "7                   1         0            1  \n",
       "8                   0         0            1  \n",
       "9                   0         0            1  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "legitimate_urls.head(10)\n",
    "phishing_urls.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data PreProcessing\n",
    "#### Data is in two data frames so we merge them to make one dataframe\n",
    "Note: two dataframes has same column names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "urls = legitimate_urls.append(phishing_urls)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>Having_@_symbol</th>\n",
       "      <th>Having_IP</th>\n",
       "      <th>Prefix_suffix_separation</th>\n",
       "      <th>Redirection_//_symbol</th>\n",
       "      <th>Sub_domains</th>\n",
       "      <th>URL_Length</th>\n",
       "      <th>age_domain</th>\n",
       "      <th>dns_record</th>\n",
       "      <th>domain_registration_length</th>\n",
       "      <th>http_tokens</th>\n",
       "      <th>label</th>\n",
       "      <th>statistical_report</th>\n",
       "      <th>tiny_url</th>\n",
       "      <th>web_traffic</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</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",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Having_@_symbol  Having_IP  Prefix_suffix_separation  \\\n",
       "0                0          0                         0   \n",
       "1                0          0                         0   \n",
       "2                0          0                         0   \n",
       "3                0          0                         1   \n",
       "4                0          0                         0   \n",
       "\n",
       "   Redirection_//_symbol  Sub_domains  URL_Length  age_domain  dns_record  \\\n",
       "0                      0            0           0           0           0   \n",
       "1                      0            0           0           0           0   \n",
       "2                      0            0           0           1           1   \n",
       "3                      0            0           0           0           0   \n",
       "4                      0            2           0           0           0   \n",
       "\n",
       "   domain_registration_length  http_tokens  label  statistical_report  \\\n",
       "0                           1            0      0                   0   \n",
       "1                           1            0      1                   0   \n",
       "2                           1            0      1                   1   \n",
       "3                           0            0      1                   0   \n",
       "4                           0            0      0                   0   \n",
       "\n",
       "   tiny_url  web_traffic  \n",
       "0         0            1  \n",
       "1         0            1  \n",
       "2         0            1  \n",
       "3         0            2  \n",
       "4         0            2  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "urls.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "14"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "urls.shape[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Domain', 'Having_@_symbol', 'Having_IP', 'Path',\n",
       "       'Prefix_suffix_separation', 'Protocol', 'Redirection_//_symbol',\n",
       "       'Sub_domains', 'URL_Length', 'age_domain', 'dns_record',\n",
       "       'domain_registration_length', 'http_tokens', 'label',\n",
       "       'statistical_report', 'tiny_url', 'web_traffic'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "urls.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Removing Unnecessary columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "urls = urls.drop(urls.columns[[0,3,5]],axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Since we merged two dataframes top 1000 rows will have legitimate urls and bottom 1000 rows will have phishing urls. So if we split the data now and create a model for it will overfit so we need to shuffle the rows before splitting the data into training set and test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# shuffling the rows in the dataset so that when splitting the train and test set are equally distributed\n",
    "urls = urls.sample(frac=1).reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Removing class variable from the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "urls_without_labels = urls.drop('label',axis=1)\n",
    "urls_without_labels.columns\n",
    "labels = urls['label']\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### splitting the data into train data and test data\n",
    "\n",
    "Dividing the data in the ratio of 70:30 [train_data:test_data]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "data_train, data_test, labels_train, labels_test = train_test_split(urls_without_labels, labels, test_size=0.30, random_state=110)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1410 605 1410 605\n"
     ]
    }
   ],
   "source": [
    "print(len(data_train),len(data_test),len(labels_train),len(labels_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### checking the split of labels in train and test data\n",
    "\n",
    "The split should be in equal proportion for both classes\n",
    "\n",
    "Phishing - 1\n",
    "\n",
    "Legitimate - 0\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    726\n",
       "1    684\n",
       "Name: label, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#initially checking the split of labels_train data \n",
    "labels_train.value_counts()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By above results it is shown that the split of both classes are almost equal!\n",
    "\n",
    "Testing the same for labels of test data also"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    314\n",
       "0    291\n",
       "Name: label, dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#checking the split for labels_test data\n",
    "labels_test.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "as the split is almost in equal proportion we can train the model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Creating the model and fitting the data into the model\n",
    "\n",
    "creating the model with default parameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "random_forest_classifier = RandomForestClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "            max_depth=None, max_features='auto', 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, n_estimators=10, n_jobs=1,\n",
       "            oob_score=False, random_state=None, verbose=0,\n",
       "            warm_start=False)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "random_forest_classifier.fit(data_train,labels_train)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Predicting the result for test data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "prediction_label = random_forest_classifier.predict(data_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Creating confusion matrix and checking the accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[256  35]\n",
      " [ 70 244]]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.8264462809917356"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix,accuracy_score\n",
    "cpnfusionMatrix = confusion_matrix(labels_test,prediction_label)\n",
    "print(cpnfusionMatrix)\n",
    "accuracy_score(labels_test,prediction_label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Improving the efficiency of model by specifying max_depth as well as number of tress "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "custom_random_forest_classifier = RandomForestClassifier(n_estimators=500, max_depth=20, max_leaf_nodes=10000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "            max_depth=20, max_features='auto', max_leaf_nodes=10000,\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, n_estimators=500, n_jobs=1,\n",
       "            oob_score=False, random_state=None, verbose=0,\n",
       "            warm_start=False)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "custom_random_forest_classifier.fit(data_train,labels_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "custom_classifier_prediction_label = custom_random_forest_classifier.predict(data_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[258  33]\n",
      " [ 72 242]]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0.8264462809917356"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#from sklearn.metrics import confusion_matrix,accuracy_score\n",
    "confusionMatrix2 = confusion_matrix(labels_test,custom_classifier_prediction_label)\n",
    "print(confusionMatrix2)\n",
    "accuracy_score(labels_test,custom_classifier_prediction_label)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Importance Plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "indices of columns : [ 5 12 10  6  7  4  8 11  2  1  0  3  9]\n",
      "\n",
      " ***Feature ranking: *** \n",
      "\n",
      "Feature name : Importance\n",
      "1 URL_Length   :  0.2146949423998274 \n",
      "\n",
      "2 web_traffic   :  0.17063818361629207 \n",
      "\n",
      "3 statistical_report   :  0.16383533750353504 \n",
      "\n",
      "4 age_domain   :  0.08935482374990723 \n",
      "\n",
      "5 dns_record   :  0.08640564469618774 \n",
      "\n",
      "6 Sub_domains   :  0.07526907894724433 \n",
      "\n",
      "7 domain_registration_length   :  0.069144732556444 \n",
      "\n",
      "8 tiny_url   :  0.058182556312364145 \n",
      "\n",
      "9 Prefix_suffix_separation   :  0.04874487888426426 \n",
      "\n",
      "10 Having_IP   :  0.009472164696868646 \n",
      "\n",
      "11 Having_@_symbol   :  0.006893716943819115 \n",
      "\n",
      "12 Redirection_//_symbol   :  0.006107522575786504 \n",
      "\n",
      "13 http_tokens   :  0.0012564171174594804 \n",
      "\n",
      "**** The blue bars are the feature importances of the randomforest classifier, along with their inter-trees variability*****\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAB9AAAARvCAYAAABATH9EAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nADoAAAAAAAAAJEn+Pxpl0l5WvLXGAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 2520x1440 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "#feature_importances_ : array of shape = [n_features] ------ The feature importances (the higher, the more important the feature).\n",
    "\n",
    "#feature_importances_  -- This method returns the quantified relative importance in the order the features were fed to the algorithm\n",
    "\n",
    "importances = custom_random_forest_classifier.feature_importances_\n",
    "\n",
    "#std = np.std([tree.feature_importances_ for tree in custom_random_forest_classifier.estimators_],axis=0)   #[[[estimators_ :explaination ---  list of DecisionTreeClassifier ----- (The collection of fitted sub-estimators.)]]]\n",
    "\n",
    "#To make the plot pretty, we’ll instead sort the features from most to least important.\n",
    "indices = np.argsort(importances)[::-1] \n",
    "print(f\"indices of columns : {indices}\")\n",
    "\n",
    "# Print the feature ranking\n",
    "print(\"\\n ***Feature ranking: *** \\n\")\n",
    "print(\"Feature name : Importance\")\n",
    "\n",
    "for f in range(data_train.shape[1]):\n",
    "    print(f\"{f+1} {data_train.columns[indices[f]]}   :  {importances[indices[f]]} \\n\")\n",
    "    \n",
    "print(\"**** The blue bars are the feature importances of the randomforest classifier, along with their inter-trees variability*****\")\n",
    "\n",
    "# Plot the feature importances of the forest\n",
    "plt.figure()\n",
    "plt.title(\"Feature importances\")\n",
    "plt.bar(range(data_train.shape[1]), importances[indices],\n",
    "       color=\"b\", align=\"center\")   \n",
    "#yerr=std[indices] -- this is another parameter that can be included if std is calculated above\n",
    "#and also it gives error bar that's the reason we calculate std above. but here we are not making it plot.\n",
    "\n",
    "plt.xticks(range(data_train.shape[1]), data_train.columns[indices])\n",
    "plt.xlim([-1, data_train.shape[1]])\n",
    "\n",
    "plt.rcParams['figure.figsize'] = (35,15)  #this will increase the size of the plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#click on the image to get clear view"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}