src/notebooks/Benign_dataSetup_explore.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt  \n",
    "# library to clean data \n",
    "import re  \n",
    "  \n",
    "# Natural Language Tool Kit \n",
    "import nltk  \n",
    "# enable flag to how all output\n",
    "from IPython.core.interactiveshell import InteractiveShell\n",
    "InteractiveShell.ast_node_interactivity = \"all\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Overview\n",
    "Here, our aim is to build a dataset that can be used to create a machine learning model to detect whether a domain is DGA or benign. In this exercise, we will only be focusing on domain( suffix, subdomain and tLDs are not included in final data set). \n",
    "This notebook explains process that we have followed to generate DGA dataset and various sources that we used to get Benign dataset.\n",
    "\n",
    "## Benign Dataset  \n",
    "Benign data set is downloaded from below sources  \n",
    "\n",
    "### 7.8mn unique data domain data is build using below data source\n",
    "* [Alex to 1 million(of 5/10/2020)](http://s3.amazonaws.com/alexa-static/top-1m.csv.zip)\n",
    "* [Cisco Umbrella Top 1 million](http://s3-us-west-1.amazonaws.com/umbrella-static/top-1m.csv.zip)\n",
    "* [The Majestic one million](http://downloads.majestic.com/majestic_million.csv)  \n",
    "* [DomCop Top 10M](https://www.domcop.com/files/top/top10milliondomains.csv.zip)  \n",
    "\n",
    "\n",
    "\n",
    "While creating dataset, we have using tldextract library to extract domain from URLs or fully qualified names. Since data is collected from many sources, duplicate data is removed.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>GlobalRank</th>\n",
       "      <th>TldRank</th>\n",
       "      <th>Domain</th>\n",
       "      <th>TLD</th>\n",
       "      <th>RefSubNets</th>\n",
       "      <th>RefIPs</th>\n",
       "      <th>IDN_Domain</th>\n",
       "      <th>IDN_TLD</th>\n",
       "      <th>PrevGlobalRank</th>\n",
       "      <th>PrevTldRank</th>\n",
       "      <th>PrevRefSubNets</th>\n",
       "      <th>PrevRefIPs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>facebook.com</td>\n",
       "      <td>com</td>\n",
       "      <td>479814</td>\n",
       "      <td>2769877</td>\n",
       "      <td>facebook.com</td>\n",
       "      <td>com</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>478233</td>\n",
       "      <td>2761833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>google.com</td>\n",
       "      <td>com</td>\n",
       "      <td>477558</td>\n",
       "      <td>2577827</td>\n",
       "      <td>google.com</td>\n",
       "      <td>com</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>475663</td>\n",
       "      <td>2566207</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>youtube.com</td>\n",
       "      <td>com</td>\n",
       "      <td>434599</td>\n",
       "      <td>2238686</td>\n",
       "      <td>youtube.com</td>\n",
       "      <td>com</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>432976</td>\n",
       "      <td>2230599</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>twitter.com</td>\n",
       "      <td>com</td>\n",
       "      <td>429238</td>\n",
       "      <td>2248255</td>\n",
       "      <td>twitter.com</td>\n",
       "      <td>com</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>427729</td>\n",
       "      <td>2241529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>instagram.com</td>\n",
       "      <td>com</td>\n",
       "      <td>331472</td>\n",
       "      <td>1548095</td>\n",
       "      <td>instagram.com</td>\n",
       "      <td>com</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "      <td>330180</td>\n",
       "      <td>1542691</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   GlobalRank  TldRank         Domain  TLD  RefSubNets   RefIPs  \\\n",
       "0           1        1   facebook.com  com      479814  2769877   \n",
       "1           2        2     google.com  com      477558  2577827   \n",
       "2           3        3    youtube.com  com      434599  2238686   \n",
       "3           4        4    twitter.com  com      429238  2248255   \n",
       "4           5        5  instagram.com  com      331472  1548095   \n",
       "\n",
       "      IDN_Domain IDN_TLD  PrevGlobalRank  PrevTldRank  PrevRefSubNets  \\\n",
       "0   facebook.com     com               1            1          478233   \n",
       "1     google.com     com               2            2          475663   \n",
       "2    youtube.com     com               3            3          432976   \n",
       "3    twitter.com     com               4            4          427729   \n",
       "4  instagram.com     com               5            5          330180   \n",
       "\n",
       "   PrevRefIPs  \n",
       "0     2761833  \n",
       "1     2566207  \n",
       "2     2230599  \n",
       "3     2241529  \n",
       "4     1542691  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "majestic_data = pd.read_csv(\"majestic_million.csv.zip\")\n",
    "majestic_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>Rank</th>\n",
       "      <th>Domain</th>\n",
       "      <th>Open Page Rank</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>fonts.googleapis.com</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>facebook.com</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>twitter.com</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>youtube.com</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>google.com</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Rank                Domain  Open Page Rank\n",
       "0     1  fonts.googleapis.com            10.0\n",
       "1     2          facebook.com            10.0\n",
       "2     3           twitter.com            10.0\n",
       "3     4           youtube.com            10.0\n",
       "4     5            google.com            10.0"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_10mn = pd.read_csv(\"top10milliondomains.csv.zip\")\n",
    "df_10mn.head()                      "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>google.com</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>youtube.com</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>tmall.com</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>baidu.com</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>qq.com</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   0            1\n",
       "0  1   google.com\n",
       "1  2  youtube.com\n",
       "2  3    tmall.com\n",
       "3  4    baidu.com\n",
       "4  5       qq.com"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1mn = pd.read_csv(\"top-1m.csv.zip\",header=None)\n",
    "df_1mn.head()  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_1mn.columns=['x','Domain']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>netflix.com</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>api-global.netflix.com</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>google.com</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>prod.netflix.com</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>push.prod.netflix.com</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   0                       1\n",
       "0  1             netflix.com\n",
       "1  2  api-global.netflix.com\n",
       "2  3              google.com\n",
       "3  4        prod.netflix.com\n",
       "4  5   push.prod.netflix.com"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_1mn1 = pd.read_csv(\"top-1m1.csv.zip\",header=None)\n",
    "df_1mn1.head() "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_1mn1.columns=['x','Domain']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_benign = pd.concat([majestic_data[['Domain']],df_1mn1[['Domain']],df_1mn[['Domain']],df_10mn[['Domain']]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_benign.drop_duplicates(keep='first',inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(11584643, 1)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_benign.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_benign.to_csv(\"elevenmn_benign.csv.gz\",compression=\"gzip\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 7 mn from rapid 7\n",
    "df_benign = pd.read_csv(\"elevenmn_benign.csv.gz\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "import tldextract\n",
    "\n",
    "def domain_extract(row):\n",
    "    ext = tldextract.extract(row.Domain)\n",
    "    return ext.domain\n",
    "\n",
    "# df_final.loc[:,'domain'] = df_final.apply(lambda row: domain_extract(row) , axis=1)\n",
    "# df_final.loc[:,'family'] ='benign'\n",
    "# df_final.loc[:,'label'] ='good'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Wall time: 6min 39s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "df_benign.loc[:,'domain'] = df_benign.apply(lambda row: domain_extract(row) , axis=1)\n",
    "df_benign.loc[:,'family'] ='benign'\n",
    "df_benign.loc[:,'label'] ='good'\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7362481, 5)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_benign.drop_duplicates(subset='domain',keep='first',inplace=True)\n",
    "df_benign.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "df_benign[['domain','family','label']].to_csv(\"alexa_cisco_onenpage_7mn_begin_dataset.csv.gz\",compression='gzip',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_final = pd.read_csv(\"alexa_cisco_onenpage_7mn_begin_dataset.csv.gz\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>Unnamed: 0</th>\n",
       "      <th>Domain</th>\n",
       "      <th>domain</th>\n",
       "      <th>family</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>facebook.com</td>\n",
       "      <td>facebook</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>google.com</td>\n",
       "      <td>google</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>youtube.com</td>\n",
       "      <td>youtube</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>twitter.com</td>\n",
       "      <td>twitter</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>instagram.com</td>\n",
       "      <td>instagram</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0         Domain     domain  family label\n",
       "0           0   facebook.com   facebook  benign  good\n",
       "1           1     google.com     google  benign  good\n",
       "2           2    youtube.com    youtube  benign  good\n",
       "3           3    twitter.com    twitter  benign  good\n",
       "4           4  instagram.com  instagram  benign  good"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "(7362481, 5)"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_benign.head()\n",
    "df_benign.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100139103, 1)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_r = df.drop_duplicates(keep='first')\n",
    "df_r.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_r.columns=['domain']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_all = pd.concat([df_r,df_final[['domain']]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_all.to_csv(\"All_Benign_107mn.csv.gz\",compression='gzip',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_rapid7 = df.sample(frac=.02,random_state=12313)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_rapid7.loc[:,'family'] = 'benign'\n",
    "df_rapid7.loc[:,'label'] = 'good'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_rapid7.columns = ['domain','family','label']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_rapid7.reset_index(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "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>family</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>facebook</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>google</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>youtube</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>twitter</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>instagram</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      domain  family label\n",
       "0   facebook  benign  good\n",
       "1     google  benign  good\n",
       "2    youtube  benign  good\n",
       "3    twitter  benign  good\n",
       "4  instagram  benign  good"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>family</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>maramiss</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>clydeantronerpottery</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>dianabryandesigns</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>owtandabout</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>snilv</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 domain  family label\n",
       "0              maramiss  benign  good\n",
       "1  clydeantronerpottery  benign  good\n",
       "2     dianabryandesigns  benign  good\n",
       "3           owtandabout  benign  good\n",
       "4                 snilv  benign  good"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_final.head()\n",
    "df_rapid7.drop(columns='index',inplace=True)\n",
    "df_rapid7.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(9776783, 3)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_benign_final = pd.concat([df_final,df_rapid7])\n",
    "df_benign_final.drop_duplicates(subset='domain',keep='first',inplace=True)\n",
    "df_benign_final.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_benign_final[['domain','family','label']].to_csv(\"alexa_cisco_onenpage_webpage_rapid7_9mn_begin_dataset.csv.gz\",compression='gzip',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "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>family</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>facebook</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>google</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>youtube</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>twitter</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>instagram</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      domain  family label\n",
       "0   facebook  benign  good\n",
       "1     google  benign  good\n",
       "2    youtube  benign  good\n",
       "3    twitter  benign  good\n",
       "4  instagram  benign  good"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# number unqiue char and lenght of domain\n",
    "df_benign_final.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_benign_final = df_benign_final[~df_benign_final.domain.isna()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(9776782, 3)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_benign_final.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Add length and number of unique char\n",
    "df_benign_final.loc[:,'uniquechar'] = df_benign_final.apply(lambda row: len(set(row.domain)) , axis=1)\n",
    "df_benign_final.loc[:,'length'] = df_benign_final.apply(lambda row: len(row.domain) , axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "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>family</th>\n",
       "      <th>label</th>\n",
       "      <th>uniquechar</th>\n",
       "      <th>length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>facebook</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>google</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>youtube</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>twitter</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "      <td>5</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>instagram</td>\n",
       "      <td>benign</td>\n",
       "      <td>good</td>\n",
       "      <td>8</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      domain  family label  uniquechar  length\n",
       "0   facebook  benign  good           7       8\n",
       "1     google  benign  good           4       6\n",
       "2    youtube  benign  good           6       7\n",
       "3    twitter  benign  good           5       7\n",
       "4  instagram  benign  good           8       9"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_benign_final.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(9776782, 5)"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_benign_final.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x00000264E719CE88>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x00000264E5A47188>]],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = (20,10)\n",
    "df_benign_final[['uniquechar','length']].hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_all[['uniquechar','length']].hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x264e7182088>"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_benign_final[['uniquechar','length']].plot.kde()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from nltk import word_tokenize, ngrams,bigrams\n",
    "VALID_CHARS = 'abcdefghijklmnopqrstuvwxyz0123456789-_.'\n",
    "TABLE = None\n",
    "#benign_bigrams = nltk.bigrams(df_benign_final.domain)\n",
    "#fdist = nltk.FreqDist(benign_bigrams)\n",
    "\n",
    "\n",
    "def generate_data_matrix(data):\n",
    "    x_label = [c for c in VALID_CHARS ]\n",
    "    y_label = [c for c in VALID_CHARS ]\n",
    "    x = [VALID_CHARS.index(c) for c in VALID_CHARS ]\n",
    "    y = [VALID_CHARS.index(c) for c in VALID_CHARS ]\n",
    "    row = [0 for _ in VALID_CHARS ]\n",
    "    MATRIX = np.array([row for _ in VALID_CHARS])\n",
    "    \n",
    "    # create ngram of var\n",
    "    for word in data :\n",
    "        try:\n",
    "            if word and word is not ' ':\n",
    "                freqdist = dict(nltk.FreqDist(bigrams(word)))\n",
    "                for key in freqdist.keys():\n",
    "                    value = freqdist[key]\n",
    "                    #print(key)\n",
    "                    xx = VALID_CHARS.index(key[0])\n",
    "                    yy = VALID_CHARS.index(key[1])\n",
    "                    # update matrix value\n",
    "                    MATRIX[xx,yy] +=value\n",
    "        except: \n",
    "            pass\n",
    "           \n",
    "    return MATRIX,x_label,y_label\n",
    "\n",
    "# generate heatmap\n",
    "def generate_heatmap(mat,xlabel,ylabel,datatype):\n",
    "    fig, ax = plt.subplots(figsize=(20,20))\n",
    "    im = ax.imshow(mat,cmap=plt.get_cmap('summer'))\n",
    "    ax.set_xticks(np.arange(len(xlabel)))\n",
    "    ax.set_yticks(np.arange(len(ylabel)))\n",
    "    ax.set_xticklabels(xlabel)\n",
    "    ax.set_yticklabels(ylabel)\n",
    "    ax.set_title(f'{datatype} Heatmap of Bi-Grams')\n",
    "# create bar\n",
    "    cbar = fig.colorbar(im,ax=ax)\n",
    "    plt.show()\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create zero matrixs\n",
    "MAT = generate_heatmap('a','s')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "MAT[1,0]\n",
    "#list(bigrams('faadadacdcrvrvrbasasaaa'))\n",
    "data = ['faadadacdcrvrvrbasasaaa','facebook','googgle']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "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",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>marketing-farmaceutico</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>redetzki</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>eleboardsports</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>superecki</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>bustalimeco</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   domain\n",
       "0  marketing-farmaceutico\n",
       "1                redetzki\n",
       "2          eleboardsports\n",
       "3               superecki\n",
       "4             bustalimeco"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_all.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "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",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>74530495</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>122</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>149790</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>398286</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         domain\n",
       "74530495    NaN\n",
       "122         NaN\n",
       "149790      NaN\n",
       "398286      NaN"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_all[df_all.domain.isna()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_all.drop(index=74530495,inplace=True)\n",
    "df_all.drop(index=122,inplace=True)\n",
    "df_all.drop(index=149790,inplace=True)\n",
    "df_all.drop(index=398286,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "mat,xlabel,ylabel = generate_data_matrix(df_all.domain)\n",
    "generate_heatmap(mat,xlabel,ylabel,'Benign 107 mn')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "mat,xlabel,ylabel = generate_data_matrix(df_benign_final.domain)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "generate_heatmap(mat,xlabel,ylabel,'Benign 9 mn')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## DGA Dataset\n",
    "9.7 millions DGA data is generate using 45 different algorithms. Please see here for more details on how data in generated. [github](<link to github readme page of dataset).\n",
    "Docker image of generation program is avaliable at [dockerhub](docker hub link)\n",
    "\n",
    "several output files are generate using above program. All files are used for creating feature data as below.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_dga_1 = pd.read_csv(\"onemillon_dga_1.csv.gz\")\n",
    "df_dga_2 = pd.read_csv(\"output_dga_2.csv.gz\")\n",
    "df_dga_3 = pd.read_csv(\"output_dga_3.csv.gz\")\n",
    "df_dga_4 = pd.read_csv(\"output_dga_4.csv.gz\")\n",
    "df_dga_5 = pd.read_csv(\"output_dga_5.csv.gz\")\n",
    "df_dga_6 = pd.read_csv(\"output_dga_6.csv.gz\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "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>domainName</th>\n",
       "      <th>family</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>kdeycqkshsoctvg8.net</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>yt1ud1mdsiaffc8n.cn</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>2cvy2esvfagmai9l.ru</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>unobf48gc6sx6476.com</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>a44z7he1iwnzs7o4.ru</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             domainName  family label\n",
       "0  kdeycqkshsoctvg8.net  chinad   bad\n",
       "1   yt1ud1mdsiaffc8n.cn  chinad   bad\n",
       "2   2cvy2esvfagmai9l.ru  chinad   bad\n",
       "3  unobf48gc6sx6476.com  chinad   bad\n",
       "4   a44z7he1iwnzs7o4.ru  chinad   bad"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dga_1.head()\n",
    "df_dga_1.columns = ['Domain','family','label']\n",
    "df_dga_2.columns = ['Domain','family','label']\n",
    "df_dga_3.columns = ['Domain','family','label']\n",
    "df_dga_4.columns = ['Domain','family','label']\n",
    "df_dga_5.columns = ['Domain','family','label']\n",
    "df_dga_6.columns = ['Domain','family','label']\n",
    "df_dga = pd.concat([df_dga_1[['Domain','family','label']],\n",
    "                    df_dga_2[['Domain','family','label']],\n",
    "                    df_dga_3[['Domain','family','label']],\n",
    "                   df_dga_4[['Domain','family','label']],\n",
    "                   df_dga_5[['Domain','family','label']],\n",
    "                   df_dga_6[['Domain','family','label']],])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(21350295, 3)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dga.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_dga.drop_duplicates(subset='Domain',keep='first',inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(12590985, 3)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dga.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_dga.drop(index=816505,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_dga.loc[:,'domain'] = df_dga.apply(lambda row: domain_extract(row) , axis=1)\n",
    "df_dga.loc[:,'uniquechar'] = df_dga.apply(lambda row: len(set(row.domain)) , axis=1)\n",
    "df_dga.loc[:,'length'] = df_dga.apply(lambda row: len(row.domain) , axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10644128, 6)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dga.drop_duplicates(subset='domain',keep='first',inplace=True)\n",
    "df_dga.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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>label</th>\n",
       "      <th>domain</th>\n",
       "      <th>uniquechar</th>\n",
       "      <th>length</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>family</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>bamital</td>\n",
       "      <td>48790</td>\n",
       "      <td>48790</td>\n",
       "      <td>48790</td>\n",
       "      <td>48790</td>\n",
       "      <td>48790</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>banjori</td>\n",
       "      <td>61488</td>\n",
       "      <td>61488</td>\n",
       "      <td>61488</td>\n",
       "      <td>61488</td>\n",
       "      <td>61488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>bobax</td>\n",
       "      <td>207999</td>\n",
       "      <td>207999</td>\n",
       "      <td>207999</td>\n",
       "      <td>207999</td>\n",
       "      <td>207999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>chinad</td>\n",
       "      <td>203776</td>\n",
       "      <td>203776</td>\n",
       "      <td>203776</td>\n",
       "      <td>203776</td>\n",
       "      <td>203776</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>corebot</td>\n",
       "      <td>817754</td>\n",
       "      <td>817754</td>\n",
       "      <td>817754</td>\n",
       "      <td>817754</td>\n",
       "      <td>817754</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>dircypt</td>\n",
       "      <td>1328</td>\n",
       "      <td>1328</td>\n",
       "      <td>1328</td>\n",
       "      <td>1328</td>\n",
       "      <td>1328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>dnscharger</td>\n",
       "      <td>207685</td>\n",
       "      <td>207685</td>\n",
       "      <td>207685</td>\n",
       "      <td>207685</td>\n",
       "      <td>207685</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>dyre</td>\n",
       "      <td>198413</td>\n",
       "      <td>198413</td>\n",
       "      <td>198413</td>\n",
       "      <td>198413</td>\n",
       "      <td>198413</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>enviserve</td>\n",
       "      <td>636</td>\n",
       "      <td>636</td>\n",
       "      <td>636</td>\n",
       "      <td>636</td>\n",
       "      <td>636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>fobber</td>\n",
       "      <td>204058</td>\n",
       "      <td>204058</td>\n",
       "      <td>204058</td>\n",
       "      <td>204058</td>\n",
       "      <td>204058</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>gameoverzeus</td>\n",
       "      <td>9377</td>\n",
       "      <td>9377</td>\n",
       "      <td>9377</td>\n",
       "      <td>9377</td>\n",
       "      <td>9377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>gozi</td>\n",
       "      <td>408492</td>\n",
       "      <td>408492</td>\n",
       "      <td>408492</td>\n",
       "      <td>408492</td>\n",
       "      <td>408492</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>kraken</td>\n",
       "      <td>12166</td>\n",
       "      <td>12166</td>\n",
       "      <td>12166</td>\n",
       "      <td>12166</td>\n",
       "      <td>12166</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>locy</td>\n",
       "      <td>428245</td>\n",
       "      <td>428245</td>\n",
       "      <td>428245</td>\n",
       "      <td>428245</td>\n",
       "      <td>428245</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>matsnu</td>\n",
       "      <td>199990</td>\n",
       "      <td>199990</td>\n",
       "      <td>199990</td>\n",
       "      <td>199990</td>\n",
       "      <td>199990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>monero</td>\n",
       "      <td>42915</td>\n",
       "      <td>42915</td>\n",
       "      <td>42915</td>\n",
       "      <td>42915</td>\n",
       "      <td>42915</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>murofet</td>\n",
       "      <td>429252</td>\n",
       "      <td>429252</td>\n",
       "      <td>429252</td>\n",
       "      <td>429252</td>\n",
       "      <td>429252</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>mydoom</td>\n",
       "      <td>201967</td>\n",
       "      <td>201967</td>\n",
       "      <td>201967</td>\n",
       "      <td>201967</td>\n",
       "      <td>201967</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>necurs</td>\n",
       "      <td>203635</td>\n",
       "      <td>203635</td>\n",
       "      <td>203635</td>\n",
       "      <td>203635</td>\n",
       "      <td>203635</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>newgoz</td>\n",
       "      <td>423988</td>\n",
       "      <td>423988</td>\n",
       "      <td>423988</td>\n",
       "      <td>423988</td>\n",
       "      <td>423988</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>nymaim</td>\n",
       "      <td>876890</td>\n",
       "      <td>876890</td>\n",
       "      <td>876890</td>\n",
       "      <td>876890</td>\n",
       "      <td>876890</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>padcrypt</td>\n",
       "      <td>670481</td>\n",
       "      <td>670481</td>\n",
       "      <td>670481</td>\n",
       "      <td>670481</td>\n",
       "      <td>670481</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>pitou</td>\n",
       "      <td>69263</td>\n",
       "      <td>69263</td>\n",
       "      <td>69263</td>\n",
       "      <td>69263</td>\n",
       "      <td>69263</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>pizd</td>\n",
       "      <td>16384</td>\n",
       "      <td>16384</td>\n",
       "      <td>16384</td>\n",
       "      <td>16384</td>\n",
       "      <td>16384</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>proslikefan</td>\n",
       "      <td>311846</td>\n",
       "      <td>311846</td>\n",
       "      <td>311846</td>\n",
       "      <td>311846</td>\n",
       "      <td>311846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>pushdo</td>\n",
       "      <td>91849</td>\n",
       "      <td>91849</td>\n",
       "      <td>91849</td>\n",
       "      <td>91849</td>\n",
       "      <td>91849</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>pykspa</td>\n",
       "      <td>483212</td>\n",
       "      <td>483212</td>\n",
       "      <td>483212</td>\n",
       "      <td>483212</td>\n",
       "      <td>483212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>qadars</td>\n",
       "      <td>15600</td>\n",
       "      <td>15600</td>\n",
       "      <td>15600</td>\n",
       "      <td>15600</td>\n",
       "      <td>15600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>qakbot</td>\n",
       "      <td>257129</td>\n",
       "      <td>257129</td>\n",
       "      <td>257129</td>\n",
       "      <td>257129</td>\n",
       "      <td>257129</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>qsnatch</td>\n",
       "      <td>231103</td>\n",
       "      <td>231103</td>\n",
       "      <td>231103</td>\n",
       "      <td>231103</td>\n",
       "      <td>231103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>ramdo</td>\n",
       "      <td>207272</td>\n",
       "      <td>207272</td>\n",
       "      <td>207272</td>\n",
       "      <td>207272</td>\n",
       "      <td>207272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>ramnit</td>\n",
       "      <td>111445</td>\n",
       "      <td>111445</td>\n",
       "      <td>111445</td>\n",
       "      <td>111445</td>\n",
       "      <td>111445</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>ranbyus</td>\n",
       "      <td>51480</td>\n",
       "      <td>51480</td>\n",
       "      <td>51480</td>\n",
       "      <td>51480</td>\n",
       "      <td>51480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>reconyc</td>\n",
       "      <td>857504</td>\n",
       "      <td>857504</td>\n",
       "      <td>857504</td>\n",
       "      <td>857504</td>\n",
       "      <td>857504</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>rovnix</td>\n",
       "      <td>411183</td>\n",
       "      <td>411183</td>\n",
       "      <td>411183</td>\n",
       "      <td>411183</td>\n",
       "      <td>411183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>shiotob</td>\n",
       "      <td>208305</td>\n",
       "      <td>208305</td>\n",
       "      <td>208305</td>\n",
       "      <td>208305</td>\n",
       "      <td>208305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>simda</td>\n",
       "      <td>11550</td>\n",
       "      <td>11550</td>\n",
       "      <td>11550</td>\n",
       "      <td>11550</td>\n",
       "      <td>11550</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>sisron</td>\n",
       "      <td>5203</td>\n",
       "      <td>5203</td>\n",
       "      <td>5203</td>\n",
       "      <td>5203</td>\n",
       "      <td>5203</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>suppobox</td>\n",
       "      <td>63023</td>\n",
       "      <td>63023</td>\n",
       "      <td>63023</td>\n",
       "      <td>63023</td>\n",
       "      <td>63023</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>symmi</td>\n",
       "      <td>205615</td>\n",
       "      <td>205615</td>\n",
       "      <td>205615</td>\n",
       "      <td>205615</td>\n",
       "      <td>205615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>tempedreve</td>\n",
       "      <td>1841</td>\n",
       "      <td>1841</td>\n",
       "      <td>1841</td>\n",
       "      <td>1841</td>\n",
       "      <td>1841</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>tinba</td>\n",
       "      <td>22004</td>\n",
       "      <td>22004</td>\n",
       "      <td>22004</td>\n",
       "      <td>22004</td>\n",
       "      <td>22004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>tofsee</td>\n",
       "      <td>208363</td>\n",
       "      <td>208363</td>\n",
       "      <td>208363</td>\n",
       "      <td>208363</td>\n",
       "      <td>208363</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>torpig</td>\n",
       "      <td>5623</td>\n",
       "      <td>5623</td>\n",
       "      <td>5623</td>\n",
       "      <td>5623</td>\n",
       "      <td>5623</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>unjavascript</td>\n",
       "      <td>205726</td>\n",
       "      <td>205726</td>\n",
       "      <td>205726</td>\n",
       "      <td>205726</td>\n",
       "      <td>205726</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>unmalware</td>\n",
       "      <td>158</td>\n",
       "      <td>158</td>\n",
       "      <td>158</td>\n",
       "      <td>158</td>\n",
       "      <td>158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>vawtrak</td>\n",
       "      <td>150817</td>\n",
       "      <td>150817</td>\n",
       "      <td>150817</td>\n",
       "      <td>150817</td>\n",
       "      <td>150817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>vidro</td>\n",
       "      <td>8916</td>\n",
       "      <td>8916</td>\n",
       "      <td>8916</td>\n",
       "      <td>8916</td>\n",
       "      <td>8916</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>zloader</td>\n",
       "      <td>572389</td>\n",
       "      <td>572389</td>\n",
       "      <td>572389</td>\n",
       "      <td>572389</td>\n",
       "      <td>572389</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Domain   label  domain  uniquechar  length\n",
       "family                                                  \n",
       "bamital        48790   48790   48790       48790   48790\n",
       "banjori        61488   61488   61488       61488   61488\n",
       "bobax         207999  207999  207999      207999  207999\n",
       "chinad        203776  203776  203776      203776  203776\n",
       "corebot       817754  817754  817754      817754  817754\n",
       "dircypt         1328    1328    1328        1328    1328\n",
       "dnscharger    207685  207685  207685      207685  207685\n",
       "dyre          198413  198413  198413      198413  198413\n",
       "enviserve        636     636     636         636     636\n",
       "fobber        204058  204058  204058      204058  204058\n",
       "gameoverzeus    9377    9377    9377        9377    9377\n",
       "gozi          408492  408492  408492      408492  408492\n",
       "kraken         12166   12166   12166       12166   12166\n",
       "locy          428245  428245  428245      428245  428245\n",
       "matsnu        199990  199990  199990      199990  199990\n",
       "monero         42915   42915   42915       42915   42915\n",
       "murofet       429252  429252  429252      429252  429252\n",
       "mydoom        201967  201967  201967      201967  201967\n",
       "necurs        203635  203635  203635      203635  203635\n",
       "newgoz        423988  423988  423988      423988  423988\n",
       "nymaim        876890  876890  876890      876890  876890\n",
       "padcrypt      670481  670481  670481      670481  670481\n",
       "pitou          69263   69263   69263       69263   69263\n",
       "pizd           16384   16384   16384       16384   16384\n",
       "proslikefan   311846  311846  311846      311846  311846\n",
       "pushdo         91849   91849   91849       91849   91849\n",
       "pykspa        483212  483212  483212      483212  483212\n",
       "qadars         15600   15600   15600       15600   15600\n",
       "qakbot        257129  257129  257129      257129  257129\n",
       "qsnatch       231103  231103  231103      231103  231103\n",
       "ramdo         207272  207272  207272      207272  207272\n",
       "ramnit        111445  111445  111445      111445  111445\n",
       "ranbyus        51480   51480   51480       51480   51480\n",
       "reconyc       857504  857504  857504      857504  857504\n",
       "rovnix        411183  411183  411183      411183  411183\n",
       "shiotob       208305  208305  208305      208305  208305\n",
       "simda          11550   11550   11550       11550   11550\n",
       "sisron          5203    5203    5203        5203    5203\n",
       "suppobox       63023   63023   63023       63023   63023\n",
       "symmi         205615  205615  205615      205615  205615\n",
       "tempedreve      1841    1841    1841        1841    1841\n",
       "tinba          22004   22004   22004       22004   22004\n",
       "tofsee        208363  208363  208363      208363  208363\n",
       "torpig          5623    5623    5623        5623    5623\n",
       "unjavascript  205726  205726  205726      205726  205726\n",
       "unmalware        158     158     158         158     158\n",
       "vawtrak       150817  150817  150817      150817  150817\n",
       "vidro           8916    8916    8916        8916    8916\n",
       "zloader       572389  572389  572389      572389  572389"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dga.groupby(['family']).count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "VALID_CHARS = 'abcdefghijklmnopqrstuvwxyz0123456789-_.'\n",
    "LOOKUP_TABLE = None\n",
    "def pad(l, content, width):\n",
    "        l.extend([content] * (width - len(l)))\n",
    "        return l\n",
    "    \n",
    "def check_validchar(domain):\n",
    "    for c in domain.lower():\n",
    "        if c not in VALID_CHARS:\n",
    "            return False\n",
    "    return True\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_dga.loc[:,'valid'] = df_dga.apply(lambda row : check_validchar(row.domain), axis=1 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10644128, 7)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "(10461440, 7)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "df_dga.shape\n",
    "df_dga = df_dga[df_dga.valid]\n",
    "df_dga.shape\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_dga.sample(frac=.75).to_csv(\"sevenmillons_dga.csv.gz\",compression='gzip',index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### DGA data set analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_dga_final=pd.read_csv(\"sevenmillons_dga.csv.gz\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(9717248, 6)"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>family</th>\n",
       "      <th>label</th>\n",
       "      <th>domain</th>\n",
       "      <th>uniquechar</th>\n",
       "      <th>length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1hodsu7i2uou8jef.cn</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "      <td>1hodsu7i2uou8jef</td>\n",
       "      <td>13</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>7gqotxvpzsg0t7r4.info</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "      <td>7gqotxvpzsg0t7r4</td>\n",
       "      <td>13</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>6jcpi4uxktum6nro.info</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "      <td>6jcpi4uxktum6nro</td>\n",
       "      <td>14</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4k2vlny8rmbbe9u8.info</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "      <td>4k2vlny8rmbbe9u8</td>\n",
       "      <td>14</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>pgvhduc5tyciwh38.ru</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "      <td>pgvhduc5tyciwh38</td>\n",
       "      <td>14</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  Domain  family label            domain  uniquechar  length\n",
       "0    1hodsu7i2uou8jef.cn  chinad   bad  1hodsu7i2uou8jef          13      16\n",
       "1  7gqotxvpzsg0t7r4.info  chinad   bad  7gqotxvpzsg0t7r4          13      16\n",
       "2  6jcpi4uxktum6nro.info  chinad   bad  6jcpi4uxktum6nro          14      16\n",
       "3  4k2vlny8rmbbe9u8.info  chinad   bad  4k2vlny8rmbbe9u8          14      16\n",
       "4    pgvhduc5tyciwh38.ru  chinad   bad  pgvhduc5tyciwh38          14      16"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dga_final.shape\n",
    "df_dga_final.head()"
   ]
  },
  {
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"2\" halign=\"left\">uniquechar</th>\n",
       "      <th colspan=\"2\" halign=\"left\">length</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>family</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>banjori</td>\n",
       "      <td>61488</td>\n",
       "      <td>13.579479</td>\n",
       "      <td>61488</td>\n",
       "      <td>22.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>chinad</td>\n",
       "      <td>203776</td>\n",
       "      <td>13.056557</td>\n",
       "      <td>203776</td>\n",
       "      <td>16.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>corebot</td>\n",
       "      <td>817753</td>\n",
       "      <td>14.779776</td>\n",
       "      <td>817753</td>\n",
       "      <td>18.994673</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>dircypt</td>\n",
       "      <td>1328</td>\n",
       "      <td>10.935994</td>\n",
       "      <td>1328</td>\n",
       "      <td>13.958584</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>dnscharger</td>\n",
       "      <td>207685</td>\n",
       "      <td>8.437124</td>\n",
       "      <td>207685</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>fobber</td>\n",
       "      <td>204058</td>\n",
       "      <td>8.434832</td>\n",
       "      <td>204058</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>gameoverzeus</td>\n",
       "      <td>9287</td>\n",
       "      <td>18.154840</td>\n",
       "      <td>9287</td>\n",
       "      <td>25.921072</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>gozi</td>\n",
       "      <td>408494</td>\n",
       "      <td>11.679351</td>\n",
       "      <td>408494</td>\n",
       "      <td>19.552060</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>kraken</td>\n",
       "      <td>12166</td>\n",
       "      <td>8.045208</td>\n",
       "      <td>12166</td>\n",
       "      <td>9.508384</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>locy</td>\n",
       "      <td>428245</td>\n",
       "      <td>9.550640</td>\n",
       "      <td>428245</td>\n",
       "      <td>11.997284</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>matsnu</td>\n",
       "      <td>199990</td>\n",
       "      <td>13.935722</td>\n",
       "      <td>199990</td>\n",
       "      <td>26.527531</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>monero</td>\n",
       "      <td>42915</td>\n",
       "      <td>9.077758</td>\n",
       "      <td>42915</td>\n",
       "      <td>13.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>murofet</td>\n",
       "      <td>429252</td>\n",
       "      <td>13.698072</td>\n",
       "      <td>429252</td>\n",
       "      <td>20.026085</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>mydoom</td>\n",
       "      <td>201967</td>\n",
       "      <td>6.515545</td>\n",
       "      <td>201967</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>necurs</td>\n",
       "      <td>203635</td>\n",
       "      <td>10.659734</td>\n",
       "      <td>203635</td>\n",
       "      <td>13.992757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>newgoz</td>\n",
       "      <td>423988</td>\n",
       "      <td>18.688916</td>\n",
       "      <td>423988</td>\n",
       "      <td>28.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>nymaim</td>\n",
       "      <td>876880</td>\n",
       "      <td>7.332000</td>\n",
       "      <td>876880</td>\n",
       "      <td>8.506050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>others</td>\n",
       "      <td>20611</td>\n",
       "      <td>8.711077</td>\n",
       "      <td>20611</td>\n",
       "      <td>11.753772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>padcrypt</td>\n",
       "      <td>670481</td>\n",
       "      <td>8.270267</td>\n",
       "      <td>670481</td>\n",
       "      <td>16.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>pitou</td>\n",
       "      <td>69263</td>\n",
       "      <td>5.821362</td>\n",
       "      <td>69263</td>\n",
       "      <td>7.999625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>pizd</td>\n",
       "      <td>16382</td>\n",
       "      <td>9.289037</td>\n",
       "      <td>16382</td>\n",
       "      <td>12.749786</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>proslikefan</td>\n",
       "      <td>311850</td>\n",
       "      <td>6.989363</td>\n",
       "      <td>311850</td>\n",
       "      <td>8.022088</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>pushdo</td>\n",
       "      <td>91849</td>\n",
       "      <td>9.059903</td>\n",
       "      <td>91849</td>\n",
       "      <td>11.836264</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>pykspa</td>\n",
       "      <td>483213</td>\n",
       "      <td>7.864300</td>\n",
       "      <td>483213</td>\n",
       "      <td>9.635544</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>qadars</td>\n",
       "      <td>15600</td>\n",
       "      <td>10.187756</td>\n",
       "      <td>15600</td>\n",
       "      <td>12.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>qakbot</td>\n",
       "      <td>257128</td>\n",
       "      <td>12.103824</td>\n",
       "      <td>257128</td>\n",
       "      <td>16.504694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>qsnatch</td>\n",
       "      <td>231090</td>\n",
       "      <td>4.071007</td>\n",
       "      <td>231090</td>\n",
       "      <td>4.282743</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>ramdo</td>\n",
       "      <td>207272</td>\n",
       "      <td>9.389204</td>\n",
       "      <td>207272</td>\n",
       "      <td>16.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>ramnit</td>\n",
       "      <td>111445</td>\n",
       "      <td>10.447073</td>\n",
       "      <td>111445</td>\n",
       "      <td>13.499771</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>ranbyus</td>\n",
       "      <td>51480</td>\n",
       "      <td>12.508761</td>\n",
       "      <td>51480</td>\n",
       "      <td>17.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>reconyc</td>\n",
       "      <td>857504</td>\n",
       "      <td>9.305446</td>\n",
       "      <td>857504</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>rovnix</td>\n",
       "      <td>411183</td>\n",
       "      <td>12.612192</td>\n",
       "      <td>411183</td>\n",
       "      <td>22.810131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>shiotob</td>\n",
       "      <td>208305</td>\n",
       "      <td>10.396131</td>\n",
       "      <td>208305</td>\n",
       "      <td>12.509647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>simda</td>\n",
       "      <td>11549</td>\n",
       "      <td>6.179409</td>\n",
       "      <td>11549</td>\n",
       "      <td>7.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>sisron</td>\n",
       "      <td>5193</td>\n",
       "      <td>8.339303</td>\n",
       "      <td>5193</td>\n",
       "      <td>12.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>symmi</td>\n",
       "      <td>1</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>tempedreve</td>\n",
       "      <td>1841</td>\n",
       "      <td>7.462792</td>\n",
       "      <td>1841</td>\n",
       "      <td>8.487235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>tinba</td>\n",
       "      <td>22004</td>\n",
       "      <td>9.160925</td>\n",
       "      <td>22004</td>\n",
       "      <td>12.000318</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>unjavascript</td>\n",
       "      <td>205733</td>\n",
       "      <td>6.985199</td>\n",
       "      <td>205733</td>\n",
       "      <td>8.015126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>unmalware</td>\n",
       "      <td>158</td>\n",
       "      <td>7.436709</td>\n",
       "      <td>158</td>\n",
       "      <td>8.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>vawtrak</td>\n",
       "      <td>150817</td>\n",
       "      <td>7.643462</td>\n",
       "      <td>150817</td>\n",
       "      <td>9.091840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>zloader</td>\n",
       "      <td>572389</td>\n",
       "      <td>13.957808</td>\n",
       "      <td>572389</td>\n",
       "      <td>20.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             uniquechar             length           \n",
       "                  count       mean   count       mean\n",
       "family                                               \n",
       "banjori           61488  13.579479   61488  22.500000\n",
       "chinad           203776  13.056557  203776  16.000000\n",
       "corebot          817753  14.779776  817753  18.994673\n",
       "dircypt            1328  10.935994    1328  13.958584\n",
       "dnscharger       207685   8.437124  207685  10.000000\n",
       "fobber           204058   8.434832  204058  10.000000\n",
       "gameoverzeus       9287  18.154840    9287  25.921072\n",
       "gozi             408494  11.679351  408494  19.552060\n",
       "kraken            12166   8.045208   12166   9.508384\n",
       "locy             428245   9.550640  428245  11.997284\n",
       "matsnu           199990  13.935722  199990  26.527531\n",
       "monero            42915   9.077758   42915  13.000000\n",
       "murofet          429252  13.698072  429252  20.026085\n",
       "mydoom           201967   6.515545  201967  10.000000\n",
       "necurs           203635  10.659734  203635  13.992757\n",
       "newgoz           423988  18.688916  423988  28.000000\n",
       "nymaim           876880   7.332000  876880   8.506050\n",
       "others            20611   8.711077   20611  11.753772\n",
       "padcrypt         670481   8.270267  670481  16.000000\n",
       "pitou             69263   5.821362   69263   7.999625\n",
       "pizd              16382   9.289037   16382  12.749786\n",
       "proslikefan      311850   6.989363  311850   8.022088\n",
       "pushdo            91849   9.059903   91849  11.836264\n",
       "pykspa           483213   7.864300  483213   9.635544\n",
       "qadars            15600  10.187756   15600  12.000000\n",
       "qakbot           257128  12.103824  257128  16.504694\n",
       "qsnatch          231090   4.071007  231090   4.282743\n",
       "ramdo            207272   9.389204  207272  16.000000\n",
       "ramnit           111445  10.447073  111445  13.499771\n",
       "ranbyus           51480  12.508761   51480  17.000000\n",
       "reconyc          857504   9.305446  857504  10.000000\n",
       "rovnix           411183  12.612192  411183  22.810131\n",
       "shiotob          208305  10.396131  208305  12.509647\n",
       "simda             11549   6.179409   11549   7.000000\n",
       "sisron             5193   8.339303    5193  12.000000\n",
       "symmi                 1   3.000000       1   4.000000\n",
       "tempedreve         1841   7.462792    1841   8.487235\n",
       "tinba             22004   9.160925   22004  12.000318\n",
       "unjavascript     205733   6.985199  205733   8.015126\n",
       "unmalware           158   7.436709     158   8.000000\n",
       "vawtrak          150817   7.643462  150817   9.091840\n",
       "zloader          572389  13.957808  572389  20.000000"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dga_final[['family','uniquechar','length']].groupby(['family']).agg(['count','mean'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x0000026481A66BC8>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x0000026481A02348>]],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams[\"figure.figsize\"] = (20,10)\n",
    "df_dga_final[['uniquechar','length']].hist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x2648097c788>"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# density graph\n",
    "df_dga_final[['uniquechar','length']].plot.kde()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "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>family</th>\n",
       "      <th>label</th>\n",
       "      <th>domain</th>\n",
       "      <th>uniquechar</th>\n",
       "      <th>length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1hodsu7i2uou8jef.cn</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "      <td>1hodsu7i2uou8jef</td>\n",
       "      <td>13</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>7gqotxvpzsg0t7r4.info</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "      <td>7gqotxvpzsg0t7r4</td>\n",
       "      <td>13</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>6jcpi4uxktum6nro.info</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "      <td>6jcpi4uxktum6nro</td>\n",
       "      <td>14</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4k2vlny8rmbbe9u8.info</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "      <td>4k2vlny8rmbbe9u8</td>\n",
       "      <td>14</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>pgvhduc5tyciwh38.ru</td>\n",
       "      <td>chinad</td>\n",
       "      <td>bad</td>\n",
       "      <td>pgvhduc5tyciwh38</td>\n",
       "      <td>14</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  Domain  family label            domain  uniquechar  length\n",
       "0    1hodsu7i2uou8jef.cn  chinad   bad  1hodsu7i2uou8jef          13      16\n",
       "1  7gqotxvpzsg0t7r4.info  chinad   bad  7gqotxvpzsg0t7r4          13      16\n",
       "2  6jcpi4uxktum6nro.info  chinad   bad  6jcpi4uxktum6nro          14      16\n",
       "3  4k2vlny8rmbbe9u8.info  chinad   bad  4k2vlny8rmbbe9u8          14      16\n",
       "4    pgvhduc5tyciwh38.ru  chinad   bad  pgvhduc5tyciwh38          14      16"
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dga_final.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def check_validchar(domain):\n",
    "    try:\n",
    "        for c in domain.lower():\n",
    "            if c not in VALID_CHARS:\n",
    "                return False\n",
    "    except:\n",
    "        print(domain)\n",
    "        return False\n",
    "    return True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "nan\n",
      "nan\n",
      "nan\n"
     ]
    }
   ],
   "source": [
    "df_dga_final.loc[:,'valid'] = df_dga_final.apply(lambda row: check_validchar(row.domain), axis=1 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>label</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>valid</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>False</td>\n",
       "      <td>182691</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>True</td>\n",
       "      <td>9534557</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         label\n",
       "valid         \n",
       "False   182691\n",
       "True   9534557"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_dga_final[['valid','label']].groupby(['valid']).count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Heatmap\n",
    "mat,xlabel,ylabel = generate_data_matrix(df_dga_final[df_dga_final.valid].domain)\n",
    "generate_heatmap(mat,xlabel,ylabel,'DGA data 9.7 mn')"
   ]
  }
 ],
 "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
}