Anomaly_based_IDS.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "hide_input": true
   },
   "source": [
    "<span style=\"font-size:3em; text-align:center\">Information System Security -</span>\n",
    "\n",
    "<span style=\"font-size:3em; text-align:center\">Anomaly-based Intrusion Detection System</span>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data is contained in 8 different CSV files, each containing different attack data at different times. So first thing we must do is merge all the data from files into one pandas DataFrame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import glob"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Saving all .csv files in folder to list.\n",
    "path = \"MachineLearningCVE/\"\n",
    "files = [file for file in glob.glob(path + \"**/*.csv\", recursive=True)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "hide_input": false,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MachineLearningCVE\\Friday-WorkingHours-Afternoon-DDos.pcap_ISCX.csv\n",
      "MachineLearningCVE\\Friday-WorkingHours-Afternoon-PortScan.pcap_ISCX.csv\n",
      "MachineLearningCVE\\Friday-WorkingHours-Morning.pcap_ISCX.csv\n",
      "MachineLearningCVE\\Monday-WorkingHours.pcap_ISCX.csv\n",
      "MachineLearningCVE\\Thursday-WorkingHours-Afternoon-Infilteration.pcap_ISCX.csv\n",
      "MachineLearningCVE\\Thursday-WorkingHours-Morning-WebAttacks.pcap_ISCX.csv\n",
      "MachineLearningCVE\\Tuesday-WorkingHours.pcap_ISCX.csv\n",
      "MachineLearningCVE\\Wednesday-workingHours.pcap_ISCX.csv\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[None, None, None, None, None, None, None, None]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "[print(f) for f in files]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\marko\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\IPython\\core\\interactiveshell.py:3249: DtypeWarning: Columns (14,15) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  if (await self.run_code(code, result,  async_=asy)):\n",
      "c:\\users\\marko\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\IPython\\core\\interactiveshell.py:3249: DtypeWarning: Columns (14) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  if (await self.run_code(code, result,  async_=asy)):\n"
     ]
    }
   ],
   "source": [
    "# Reading all the csv files into dataframes and putting thoose DFs to one list.\n",
    "\n",
    "dataset = [pd.read_csv(f) for f in files]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(225745, 79)\n",
      "(286467, 79)\n",
      "(191033, 79)\n",
      "(529918, 79)\n",
      "(288602, 79)\n",
      "(170366, 79)\n",
      "(445909, 79)\n",
      "(692703, 79)\n"
     ]
    }
   ],
   "source": [
    "# Here we can see the number of rows and columns for each table.\n",
    "\n",
    "for d in dataset:\n",
    "    print(d.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True,  True,  True,\n",
       "        True,  True,  True,  True,  True,  True,  True])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We already established that all tables have the same number of columns, but are they the same columns?\n",
    "# This next piece of code loops over all given tables and compares each of them to all others.\n",
    "\n",
    "for i in range(0,len(dataset)):\n",
    "    if i != len(dataset)-1:\n",
    "        same_columns = dataset[i].columns == dataset[i+1].columns\n",
    "        \n",
    "        if False in same_columns:\n",
    "            print(i)\n",
    "            break\n",
    "\n",
    "same_columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Combining all tables into one dataset. This is possilbe since all tables have the same columns,\n",
    "# as we checked in the cell above.\n",
    "\n",
    "dataset = pd.concat([d for d in dataset]).drop_duplicates(keep=False)\n",
    "dataset.reset_index(drop=True, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2451998, 79)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# By checking the shape of dataset we can confirm that concatenation has been successfull.\n",
    "\n",
    "dataset.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# Preliminary data analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Some general info about the dataset. It contains roughly 2.5 million records across 79 columns. Data consists of mostly int64 and float64 types, except 3 attributes of 'object' type. \n",
    "\n",
    "Dataset contains of network traffic data during different attacks, represented with values like: port numbers, IP adressses, packet lenghts, SYN/ACK/FIN/.. flag counts, packet size and other..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true,
    "hidden": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\marko\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\IPython\\core\\interactiveshell.py:3058: DtypeWarning: Columns (15,16) have mixed types. Specify dtype option on import or set low_memory=False.\n",
      "  interactivity=interactivity, compiler=compiler, result=result)\n",
      "c:\\users\\marko\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\numpy\\lib\\arraysetops.py:568: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n",
      "  mask |= (ar1 == a)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 2451998 entries, 0 to 2451997\n",
      "Data columns (total 79 columns):\n",
      "DestinationPort            int64\n",
      "FlowDuration               int64\n",
      "TotalFwdPackets            int64\n",
      "TotalBackwardPackets       int64\n",
      "TotalLengthofFwdPackets    int64\n",
      "TotalLengthofBwdPackets    int64\n",
      "FwdPacketLengthMax         int64\n",
      "FwdPacketLengthMin         int64\n",
      "FwdPacketLengthMean        float64\n",
      "FwdPacketLengthStd         float64\n",
      "BwdPacketLengthMax         int64\n",
      "BwdPacketLengthMin         int64\n",
      "BwdPacketLengthMean        float64\n",
      "BwdPacketLengthStd         float64\n",
      "FlowBytes/s                object\n",
      "FlowPackets/s              object\n",
      "FlowIATMean                float64\n",
      "FlowIATStd                 float64\n",
      "FlowIATMax                 int64\n",
      "FlowIATMin                 int64\n",
      "FwdIATTotal                int64\n",
      "FwdIATMean                 float64\n",
      "FwdIATStd                  float64\n",
      "FwdIATMax                  int64\n",
      "FwdIATMin                  int64\n",
      "BwdIATTotal                int64\n",
      "BwdIATMean                 float64\n",
      "BwdIATStd                  float64\n",
      "BwdIATMax                  int64\n",
      "BwdIATMin                  int64\n",
      "FwdPSHFlags                int64\n",
      "BwdPSHFlags                int64\n",
      "FwdURGFlags                int64\n",
      "BwdURGFlags                int64\n",
      "FwdHeaderLength            int64\n",
      "BwdHeaderLength            int64\n",
      "FwdPackets/s               float64\n",
      "BwdPackets/s               float64\n",
      "MinPacketLength            int64\n",
      "MaxPacketLength            int64\n",
      "PacketLengthMean           float64\n",
      "PacketLengthStd            float64\n",
      "PacketLengthVariance       float64\n",
      "FINFlagCount               int64\n",
      "SYNFlagCount               int64\n",
      "RSTFlagCount               int64\n",
      "PSHFlagCount               int64\n",
      "ACKFlagCount               int64\n",
      "URGFlagCount               int64\n",
      "CWEFlagCount               int64\n",
      "ECEFlagCount               int64\n",
      "Down/UpRatio               int64\n",
      "AveragePacketSize          float64\n",
      "AvgFwdSegmentSize          float64\n",
      "AvgBwdSegmentSize          float64\n",
      "FwdHeaderLength.1          int64\n",
      "FwdAvgBytes/Bulk           int64\n",
      "FwdAvgPackets/Bulk         int64\n",
      "FwdAvgBulkRate             int64\n",
      "BwdAvgBytes/Bulk           int64\n",
      "BwdAvgPackets/Bulk         int64\n",
      "BwdAvgBulkRate             int64\n",
      "SubflowFwdPackets          int64\n",
      "SubflowFwdBytes            int64\n",
      "SubflowBwdPackets          int64\n",
      "SubflowBwdBytes            int64\n",
      "Init_Win_bytes_forward     int64\n",
      "Init_Win_bytes_backward    int64\n",
      "act_data_pkt_fwd           int64\n",
      "min_seg_size_forward       int64\n",
      "ActiveMean                 float64\n",
      "ActiveStd                  float64\n",
      "ActiveMax                  int64\n",
      "ActiveMin                  int64\n",
      "IdleMean                   float64\n",
      "IdleStd                    float64\n",
      "IdleMax                    int64\n",
      "IdleMin                    int64\n",
      "Label                      object\n",
      "dtypes: float64(22), int64(54), object(3)\n",
      "memory usage: 1.5+ GB\n"
     ]
    }
   ],
   "source": [
    "#dataset = pd.read_csv('Dataset_clean.csv', index_col=[0])\n",
    "dataset.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true,
    "hidden": true
   },
   "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>DestinationPort</th>\n",
       "      <th>FlowDuration</th>\n",
       "      <th>TotalFwdPackets</th>\n",
       "      <th>TotalBackwardPackets</th>\n",
       "      <th>TotalLengthofFwdPackets</th>\n",
       "      <th>TotalLengthofBwdPackets</th>\n",
       "      <th>FwdPacketLengthMax</th>\n",
       "      <th>FwdPacketLengthMin</th>\n",
       "      <th>FwdPacketLengthMean</th>\n",
       "      <th>FwdPacketLengthStd</th>\n",
       "      <th>...</th>\n",
       "      <th>act_data_pkt_fwd</th>\n",
       "      <th>min_seg_size_forward</th>\n",
       "      <th>ActiveMean</th>\n",
       "      <th>ActiveStd</th>\n",
       "      <th>ActiveMax</th>\n",
       "      <th>ActiveMin</th>\n",
       "      <th>IdleMean</th>\n",
       "      <th>IdleStd</th>\n",
       "      <th>IdleMax</th>\n",
       "      <th>IdleMin</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>...</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "      <td>2.451998e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>8.804922e+03</td>\n",
       "      <td>1.704883e+07</td>\n",
       "      <td>1.052762e+01</td>\n",
       "      <td>1.186615e+01</td>\n",
       "      <td>6.281065e+02</td>\n",
       "      <td>1.865192e+04</td>\n",
       "      <td>2.372695e+02</td>\n",
       "      <td>1.931056e+01</td>\n",
       "      <td>6.484734e+01</td>\n",
       "      <td>7.948213e+01</td>\n",
       "      <td>...</td>\n",
       "      <td>6.165070e+00</td>\n",
       "      <td>-3.169526e+03</td>\n",
       "      <td>9.414805e+04</td>\n",
       "      <td>4.748785e+04</td>\n",
       "      <td>1.768437e+05</td>\n",
       "      <td>6.730041e+04</td>\n",
       "      <td>9.598579e+06</td>\n",
       "      <td>5.816696e+05</td>\n",
       "      <td>1.003695e+07</td>\n",
       "      <td>9.141405e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.918522e+04</td>\n",
       "      <td>3.560952e+07</td>\n",
       "      <td>8.054868e+02</td>\n",
       "      <td>1.071646e+03</td>\n",
       "      <td>1.073530e+04</td>\n",
       "      <td>2.431588e+06</td>\n",
       "      <td>7.660169e+02</td>\n",
       "      <td>6.135372e+01</td>\n",
       "      <td>1.980198e+02</td>\n",
       "      <td>3.007317e+02</td>\n",
       "      <td>...</td>\n",
       "      <td>6.838105e+02</td>\n",
       "      <td>1.165777e+06</td>\n",
       "      <td>6.960432e+05</td>\n",
       "      <td>4.223156e+05</td>\n",
       "      <td>1.100308e+06</td>\n",
       "      <td>6.195736e+05</td>\n",
       "      <td>2.514366e+07</td>\n",
       "      <td>4.941143e+06</td>\n",
       "      <td>2.592067e+07</td>\n",
       "      <td>2.487733e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>-1.300000e+01</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>-5.368707e+08</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.300000e+01</td>\n",
       "      <td>2.270000e+02</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>2.400000e+01</td>\n",
       "      <td>6.000000e+00</td>\n",
       "      <td>6.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>6.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>2.000000e+01</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>8.000000e+01</td>\n",
       "      <td>5.978800e+04</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>6.800000e+01</td>\n",
       "      <td>1.610000e+02</td>\n",
       "      <td>4.100000e+01</td>\n",
       "      <td>2.000000e+00</td>\n",
       "      <td>3.700000e+01</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>2.000000e+01</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>4.430000e+02</td>\n",
       "      <td>5.496288e+06</td>\n",
       "      <td>6.000000e+00</td>\n",
       "      <td>5.000000e+00</td>\n",
       "      <td>3.430000e+02</td>\n",
       "      <td>1.404750e+03</td>\n",
       "      <td>2.160000e+02</td>\n",
       "      <td>3.700000e+01</td>\n",
       "      <td>5.300000e+01</td>\n",
       "      <td>7.776586e+01</td>\n",
       "      <td>...</td>\n",
       "      <td>3.000000e+00</td>\n",
       "      <td>3.200000e+01</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>6.553500e+04</td>\n",
       "      <td>1.200000e+08</td>\n",
       "      <td>2.197590e+05</td>\n",
       "      <td>2.919220e+05</td>\n",
       "      <td>1.290000e+07</td>\n",
       "      <td>6.554530e+08</td>\n",
       "      <td>2.482000e+04</td>\n",
       "      <td>2.325000e+03</td>\n",
       "      <td>5.940857e+03</td>\n",
       "      <td>7.125597e+03</td>\n",
       "      <td>...</td>\n",
       "      <td>2.135570e+05</td>\n",
       "      <td>1.380000e+02</td>\n",
       "      <td>1.100000e+08</td>\n",
       "      <td>7.420000e+07</td>\n",
       "      <td>1.100000e+08</td>\n",
       "      <td>1.100000e+08</td>\n",
       "      <td>1.200000e+08</td>\n",
       "      <td>7.690000e+07</td>\n",
       "      <td>1.200000e+08</td>\n",
       "      <td>1.200000e+08</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 76 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       DestinationPort  FlowDuration  TotalFwdPackets  TotalBackwardPackets  \\\n",
       "count     2.451998e+06  2.451998e+06     2.451998e+06          2.451998e+06   \n",
       "mean      8.804922e+03  1.704883e+07     1.052762e+01          1.186615e+01   \n",
       "std       1.918522e+04  3.560952e+07     8.054868e+02          1.071646e+03   \n",
       "min       0.000000e+00 -1.300000e+01     1.000000e+00          0.000000e+00   \n",
       "25%       5.300000e+01  2.270000e+02     2.000000e+00          1.000000e+00   \n",
       "50%       8.000000e+01  5.978800e+04     2.000000e+00          2.000000e+00   \n",
       "75%       4.430000e+02  5.496288e+06     6.000000e+00          5.000000e+00   \n",
       "max       6.553500e+04  1.200000e+08     2.197590e+05          2.919220e+05   \n",
       "\n",
       "       TotalLengthofFwdPackets  TotalLengthofBwdPackets  FwdPacketLengthMax  \\\n",
       "count             2.451998e+06             2.451998e+06        2.451998e+06   \n",
       "mean              6.281065e+02             1.865192e+04        2.372695e+02   \n",
       "std               1.073530e+04             2.431588e+06        7.660169e+02   \n",
       "min               0.000000e+00             0.000000e+00        0.000000e+00   \n",
       "25%               2.400000e+01             6.000000e+00        6.000000e+00   \n",
       "50%               6.800000e+01             1.610000e+02        4.100000e+01   \n",
       "75%               3.430000e+02             1.404750e+03        2.160000e+02   \n",
       "max               1.290000e+07             6.554530e+08        2.482000e+04   \n",
       "\n",
       "       FwdPacketLengthMin  FwdPacketLengthMean  FwdPacketLengthStd  ...  \\\n",
       "count        2.451998e+06         2.451998e+06        2.451998e+06  ...   \n",
       "mean         1.931056e+01         6.484734e+01        7.948213e+01  ...   \n",
       "std          6.135372e+01         1.980198e+02        3.007317e+02  ...   \n",
       "min          0.000000e+00         0.000000e+00        0.000000e+00  ...   \n",
       "25%          0.000000e+00         6.000000e+00        0.000000e+00  ...   \n",
       "50%          2.000000e+00         3.700000e+01        0.000000e+00  ...   \n",
       "75%          3.700000e+01         5.300000e+01        7.776586e+01  ...   \n",
       "max          2.325000e+03         5.940857e+03        7.125597e+03  ...   \n",
       "\n",
       "       act_data_pkt_fwd  min_seg_size_forward    ActiveMean     ActiveStd  \\\n",
       "count      2.451998e+06          2.451998e+06  2.451998e+06  2.451998e+06   \n",
       "mean       6.165070e+00         -3.169526e+03  9.414805e+04  4.748785e+04   \n",
       "std        6.838105e+02          1.165777e+06  6.960432e+05  4.223156e+05   \n",
       "min        0.000000e+00         -5.368707e+08  0.000000e+00  0.000000e+00   \n",
       "25%        0.000000e+00          2.000000e+01  0.000000e+00  0.000000e+00   \n",
       "50%        1.000000e+00          2.000000e+01  0.000000e+00  0.000000e+00   \n",
       "75%        3.000000e+00          3.200000e+01  0.000000e+00  0.000000e+00   \n",
       "max        2.135570e+05          1.380000e+02  1.100000e+08  7.420000e+07   \n",
       "\n",
       "          ActiveMax     ActiveMin      IdleMean       IdleStd       IdleMax  \\\n",
       "count  2.451998e+06  2.451998e+06  2.451998e+06  2.451998e+06  2.451998e+06   \n",
       "mean   1.768437e+05  6.730041e+04  9.598579e+06  5.816696e+05  1.003695e+07   \n",
       "std    1.100308e+06  6.195736e+05  2.514366e+07  4.941143e+06  2.592067e+07   \n",
       "min    0.000000e+00  0.000000e+00  0.000000e+00  0.000000e+00  0.000000e+00   \n",
       "25%    0.000000e+00  0.000000e+00  0.000000e+00  0.000000e+00  0.000000e+00   \n",
       "50%    0.000000e+00  0.000000e+00  0.000000e+00  0.000000e+00  0.000000e+00   \n",
       "75%    0.000000e+00  0.000000e+00  0.000000e+00  0.000000e+00  0.000000e+00   \n",
       "max    1.100000e+08  1.100000e+08  1.200000e+08  7.690000e+07  1.200000e+08   \n",
       "\n",
       "            IdleMin  \n",
       "count  2.451998e+06  \n",
       "mean   9.141405e+06  \n",
       "std    2.487733e+07  \n",
       "min    0.000000e+00  \n",
       "25%    0.000000e+00  \n",
       "50%    0.000000e+00  \n",
       "75%    0.000000e+00  \n",
       "max    1.200000e+08  \n",
       "\n",
       "[8 rows x 76 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Upon further inspection we can see that dataset contains 15 labels. Labels represent network/web attacks and BENIGN state which is the network traffic during normal business day. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "hidden": true,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['BENIGN' 'DDoS' 'PortScan' 'Bot' 'Infiltration' 'Web_Attack_Brute_Force'\n",
      " 'Web_Attack_XSS' 'Web_Attack_Sql_Injection' 'FTPPatator' 'SSHPatator'\n",
      " 'DoS_slowloris' 'DoS_Slowhttptest' 'DoS_Hulk' 'DoS_GoldenEye'\n",
      " 'Heartbleed']\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "15"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Dataset conatains 15 labels.\n",
    "#print(dataset[' Label'].unique())\n",
    "#len(dataset[' Label'].unique())\n",
    "\n",
    "print(dataset['Label'].unique())\n",
    "len(dataset['Label'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "hidden": true
   },
   "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>DestinationPort</th>\n",
       "      <th>FlowDuration</th>\n",
       "      <th>TotalFwdPackets</th>\n",
       "      <th>TotalBackwardPackets</th>\n",
       "      <th>TotalLengthofFwdPackets</th>\n",
       "      <th>TotalLengthofBwdPackets</th>\n",
       "      <th>FwdPacketLengthMax</th>\n",
       "      <th>FwdPacketLengthMin</th>\n",
       "      <th>FwdPacketLengthMean</th>\n",
       "      <th>FwdPacketLengthStd</th>\n",
       "      <th>...</th>\n",
       "      <th>min_seg_size_forward</th>\n",
       "      <th>ActiveMean</th>\n",
       "      <th>ActiveStd</th>\n",
       "      <th>ActiveMax</th>\n",
       "      <th>ActiveMin</th>\n",
       "      <th>IdleMean</th>\n",
       "      <th>IdleStd</th>\n",
       "      <th>IdleMax</th>\n",
       "      <th>IdleMin</th>\n",
       "      <th>Label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>54865</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>55054</td>\n",
       "      <td>109</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>55055</td>\n",
       "      <td>52</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>46236</td>\n",
       "      <td>34</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>54863</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 79 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   DestinationPort  FlowDuration  TotalFwdPackets  TotalBackwardPackets  \\\n",
       "0            54865             3                2                     0   \n",
       "1            55054           109                1                     1   \n",
       "2            55055            52                1                     1   \n",
       "3            46236            34                1                     1   \n",
       "4            54863             3                2                     0   \n",
       "\n",
       "   TotalLengthofFwdPackets  TotalLengthofBwdPackets  FwdPacketLengthMax  \\\n",
       "0                       12                        0                   6   \n",
       "1                        6                        6                   6   \n",
       "2                        6                        6                   6   \n",
       "3                        6                        6                   6   \n",
       "4                       12                        0                   6   \n",
       "\n",
       "   FwdPacketLengthMin  FwdPacketLengthMean  FwdPacketLengthStd  ...  \\\n",
       "0                   6                  6.0                 0.0  ...   \n",
       "1                   6                  6.0                 0.0  ...   \n",
       "2                   6                  6.0                 0.0  ...   \n",
       "3                   6                  6.0                 0.0  ...   \n",
       "4                   6                  6.0                 0.0  ...   \n",
       "\n",
       "   min_seg_size_forward  ActiveMean  ActiveStd  ActiveMax ActiveMin IdleMean  \\\n",
       "0                    20         0.0        0.0          0         0      0.0   \n",
       "1                    20         0.0        0.0          0         0      0.0   \n",
       "2                    20         0.0        0.0          0         0      0.0   \n",
       "3                    20         0.0        0.0          0         0      0.0   \n",
       "4                    20         0.0        0.0          0         0      0.0   \n",
       "\n",
       "   IdleStd  IdleMax  IdleMin   Label  \n",
       "0      0.0        0        0  BENIGN  \n",
       "1      0.0        0        0  BENIGN  \n",
       "2      0.0        0        0  BENIGN  \n",
       "3      0.0        0        0  BENIGN  \n",
       "4      0.0        0        0  BENIGN  \n",
       "\n",
       "[5 rows x 79 columns]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Most records in the dataset are of DDos and DOS Hulk attacks. This might pose a problem later in model training, considering that there is a very small amount of data for most attacks. Model selection will be greatly influenced by this information."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "data = dataset['Label'].where(dataset['Label'] != \"BENIGN\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "hidden": true,
    "hide_input": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13]),\n",
       " <a list of 14 Text xticklabel objects>)"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,6))\n",
    "chart = sns.countplot(data, palette=\"Set1\")\n",
    "plt.xticks(rotation=45, horizontalalignment=\"right\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# Data Cleaning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "This chapter contains data cleaning code. We go through the process of renaming columns, removing NaN and non-finite values (-inf, inf) to get the data ready for visualization and model training."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "## Renaming columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "hidden": true
   },
   "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>DestinationPort</th>\n",
       "      <th>FlowDuration</th>\n",
       "      <th>TotalFwdPackets</th>\n",
       "      <th>TotalBackwardPackets</th>\n",
       "      <th>TotalLengthofFwdPackets</th>\n",
       "      <th>TotalLengthofBwdPackets</th>\n",
       "      <th>FwdPacketLengthMax</th>\n",
       "      <th>FwdPacketLengthMin</th>\n",
       "      <th>FwdPacketLengthMean</th>\n",
       "      <th>FwdPacketLengthStd</th>\n",
       "      <th>...</th>\n",
       "      <th>min_seg_size_forward</th>\n",
       "      <th>ActiveMean</th>\n",
       "      <th>ActiveStd</th>\n",
       "      <th>ActiveMax</th>\n",
       "      <th>ActiveMin</th>\n",
       "      <th>IdleMean</th>\n",
       "      <th>IdleStd</th>\n",
       "      <th>IdleMax</th>\n",
       "      <th>IdleMin</th>\n",
       "      <th>Label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>54865</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>55054</td>\n",
       "      <td>109</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>55055</td>\n",
       "      <td>52</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>46236</td>\n",
       "      <td>34</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>54863</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 79 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   DestinationPort  FlowDuration  TotalFwdPackets  TotalBackwardPackets  \\\n",
       "0            54865             3                2                     0   \n",
       "1            55054           109                1                     1   \n",
       "2            55055            52                1                     1   \n",
       "3            46236            34                1                     1   \n",
       "4            54863             3                2                     0   \n",
       "\n",
       "   TotalLengthofFwdPackets  TotalLengthofBwdPackets  FwdPacketLengthMax  \\\n",
       "0                       12                        0                   6   \n",
       "1                        6                        6                   6   \n",
       "2                        6                        6                   6   \n",
       "3                        6                        6                   6   \n",
       "4                       12                        0                   6   \n",
       "\n",
       "   FwdPacketLengthMin  FwdPacketLengthMean  FwdPacketLengthStd  ...  \\\n",
       "0                   6                  6.0                 0.0  ...   \n",
       "1                   6                  6.0                 0.0  ...   \n",
       "2                   6                  6.0                 0.0  ...   \n",
       "3                   6                  6.0                 0.0  ...   \n",
       "4                   6                  6.0                 0.0  ...   \n",
       "\n",
       "   min_seg_size_forward  ActiveMean  ActiveStd  ActiveMax ActiveMin IdleMean  \\\n",
       "0                    20         0.0        0.0          0         0      0.0   \n",
       "1                    20         0.0        0.0          0         0      0.0   \n",
       "2                    20         0.0        0.0          0         0      0.0   \n",
       "3                    20         0.0        0.0          0         0      0.0   \n",
       "4                    20         0.0        0.0          0         0      0.0   \n",
       "\n",
       "   IdleStd  IdleMax  IdleMin   Label  \n",
       "0      0.0        0        0  BENIGN  \n",
       "1      0.0        0        0  BENIGN  \n",
       "2      0.0        0        0  BENIGN  \n",
       "3      0.0        0        0  BENIGN  \n",
       "4      0.0        0        0  BENIGN  \n",
       "\n",
       "[5 rows x 79 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Removing whitespaces in column names.\n",
    "\n",
    "col_names = [col.replace(' ', '') for col in dataset.columns]\n",
    "dataset.columns = col_names\n",
    "dataset.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['BENIGN', 'DDoS', 'PortScan', 'Bot', 'Infiltration',\n",
       "       'Web Attack � Brute Force', 'Web Attack � XSS',\n",
       "       'Web Attack � Sql Injection', 'FTP-Patator', 'SSH-Patator',\n",
       "       'DoS slowloris', 'DoS Slowhttptest', 'DoS Hulk', 'DoS GoldenEye',\n",
       "       'Heartbleed'], dtype=object)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Here we can see that 'Label' column contains some wierd characters. \n",
    "\n",
    "dataset[\"Label\"].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(['BENIGN',\n",
       "  'DDoS',\n",
       "  'PortScan',\n",
       "  'Bot',\n",
       "  'Infiltration',\n",
       "  'Web_Attack_Brute_Force',\n",
       "  'Web_Attack_XSS',\n",
       "  'Web_Attack_Sql_Injection',\n",
       "  'FTPPatator',\n",
       "  'SSHPatator',\n",
       "  'DoS_slowloris',\n",
       "  'DoS_Slowhttptest',\n",
       "  'DoS_Hulk',\n",
       "  'DoS_GoldenEye',\n",
       "  'Heartbleed'],\n",
       " 15)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# This next snippet uses regular expressions to replace wierd characters with dunders.\n",
    "\n",
    "label_names = dataset['Label'].unique()\n",
    "\n",
    "\n",
    "import re\n",
    "\n",
    "label_names = [re.sub(\"[^a-zA-Z ]+\", \"\", l) for l in label_names]\n",
    "label_names = [re.sub(\"[\\s\\s]\", '_', l) for l in label_names]\n",
    "label_names = [lab.replace(\"__\", \"_\") for lab in label_names]\n",
    "\n",
    "label_names, len(label_names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['BENIGN', 'DDoS', 'PortScan', 'Bot', 'Infiltration',\n",
       "       'Web_Attack_Brute_Force', 'Web_Attack_XSS',\n",
       "       'Web_Attack_Sql_Injection', 'FTPPatator', 'SSHPatator',\n",
       "       'DoS_slowloris', 'DoS_Slowhttptest', 'DoS_Hulk', 'DoS_GoldenEye',\n",
       "       'Heartbleed'], dtype=object)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Replacing 'Label' column values with new readable values.\n",
    "\n",
    "labels = dataset['Label'].unique()\n",
    "\n",
    "for i in range(0,len(label_names)):\n",
    "    dataset['Label'] = dataset['Label'].replace({labels[i] : label_names[i]})\n",
    "    \n",
    "dataset['Label'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "15"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(dataset['Label'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# Saving cleaned dataset.\n",
    "\n",
    "#dataset.to_csv(\"Dataset_clean.csv\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "## Removing NULL values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "hidden": true
   },
   "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>DestinationPort</th>\n",
       "      <th>FlowDuration</th>\n",
       "      <th>TotalFwdPackets</th>\n",
       "      <th>TotalBackwardPackets</th>\n",
       "      <th>TotalLengthofFwdPackets</th>\n",
       "      <th>TotalLengthofBwdPackets</th>\n",
       "      <th>FwdPacketLengthMax</th>\n",
       "      <th>FwdPacketLengthMin</th>\n",
       "      <th>FwdPacketLengthMean</th>\n",
       "      <th>FwdPacketLengthStd</th>\n",
       "      <th>...</th>\n",
       "      <th>min_seg_size_forward</th>\n",
       "      <th>ActiveMean</th>\n",
       "      <th>ActiveStd</th>\n",
       "      <th>ActiveMax</th>\n",
       "      <th>ActiveMin</th>\n",
       "      <th>IdleMean</th>\n",
       "      <th>IdleStd</th>\n",
       "      <th>IdleMax</th>\n",
       "      <th>IdleMin</th>\n",
       "      <th>Label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>54865</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>55054</td>\n",
       "      <td>109</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>55055</td>\n",
       "      <td>52</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>46236</td>\n",
       "      <td>34</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>54863</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 79 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   DestinationPort  FlowDuration  TotalFwdPackets  TotalBackwardPackets  \\\n",
       "0            54865             3                2                     0   \n",
       "1            55054           109                1                     1   \n",
       "2            55055            52                1                     1   \n",
       "3            46236            34                1                     1   \n",
       "4            54863             3                2                     0   \n",
       "\n",
       "   TotalLengthofFwdPackets  TotalLengthofBwdPackets  FwdPacketLengthMax  \\\n",
       "0                       12                        0                   6   \n",
       "1                        6                        6                   6   \n",
       "2                        6                        6                   6   \n",
       "3                        6                        6                   6   \n",
       "4                       12                        0                   6   \n",
       "\n",
       "   FwdPacketLengthMin  FwdPacketLengthMean  FwdPacketLengthStd  ...  \\\n",
       "0                   6                  6.0                 0.0  ...   \n",
       "1                   6                  6.0                 0.0  ...   \n",
       "2                   6                  6.0                 0.0  ...   \n",
       "3                   6                  6.0                 0.0  ...   \n",
       "4                   6                  6.0                 0.0  ...   \n",
       "\n",
       "   min_seg_size_forward  ActiveMean  ActiveStd  ActiveMax ActiveMin IdleMean  \\\n",
       "0                    20         0.0        0.0          0         0      0.0   \n",
       "1                    20         0.0        0.0          0         0      0.0   \n",
       "2                    20         0.0        0.0          0         0      0.0   \n",
       "3                    20         0.0        0.0          0         0      0.0   \n",
       "4                    20         0.0        0.0          0         0      0.0   \n",
       "\n",
       "   IdleStd  IdleMax  IdleMin   Label  \n",
       "0      0.0        0        0  BENIGN  \n",
       "1      0.0        0        0  BENIGN  \n",
       "2      0.0        0        0  BENIGN  \n",
       "3      0.0        0        0  BENIGN  \n",
       "4      0.0        0        0  BENIGN  \n",
       "\n",
       "[5 rows x 79 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#dataset = pd.read_csv(\"Dataset_clean.csv\", index_col=0)\n",
    "dataset.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking if there are any NULL values in the dataset.\n",
    "\n",
    "dataset.isnull().values.any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['FlowBytes/s']"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking which column/s contain NULL values.\n",
    "\n",
    "[col for col in dataset if dataset[col].isnull().values.any()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "334"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking how many NULL values it this column contains.\n",
    "\n",
    "dataset['FlowBytes/s'].isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.013621544552646454"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Considering that only 334 rows contain NULL vlaues in the entire dataset, which makes about 0.01%, we\n",
    "# can safely remove all NULL rows without spoiling the data.\n",
    "\n",
    "334/dataset.shape[0]*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "334"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Removing rows that contain NULL values and checking if number of removed rows is equal to the number of null values.\n",
    "\n",
    "before = dataset.shape\n",
    "\n",
    "dataset.dropna(inplace=True)\n",
    "\n",
    "after = dataset.shape\n",
    "\n",
    "before[0] - after[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.isnull().any().any()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "## Removing *non-finite* values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "labl = dataset['Label']\n",
    "dataset = dataset.loc[:, dataset.columns != 'Label'].astype('float64')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking if all values are finite.\n",
    "\n",
    "np.all(np.isfinite(dataset))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['FlowBytes/s', 'FlowPackets/s']"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking what column/s contain non-finite values.\n",
    "\n",
    "nonfinite = [col for col in dataset if not np.all(np.isfinite(dataset[col]))]\n",
    "\n",
    "nonfinite"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1132"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking how many non-finite values each column contains.\n",
    "\n",
    "finite = np.isfinite(dataset['FlowBytes/s']).sum()\n",
    "\n",
    "dataset.shape[0] - finite"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1132"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking how many non-finite values each column contains.\n",
    "\n",
    "finite = np.isfinite(dataset['FlowPackets/s']).sum()\n",
    "\n",
    "dataset.shape[0] - finite"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# Same as before, since there is a small number of non-finite values we can safely remove them from the dataset\n",
    "# without spoiling the dataset.\n",
    "\n",
    "# Replacing infinite values with NaN values.\n",
    "dataset = dataset.replace([np.inf, -np.inf], np.nan)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We can see that now we have Nan values again.\n",
    "\n",
    "np.any(np.isnan(dataset))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# Bringing the Labels back into the dataset before deliting Nan rows.\n",
    "\n",
    "dataset = dataset.merge(labl, how='outer', left_index=True, right_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# Removing new NaN values.\n",
    "\n",
    "dataset.dropna(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2450532, 79)"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "hidden": true
   },
   "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>DestinationPort</th>\n",
       "      <th>FlowDuration</th>\n",
       "      <th>TotalFwdPackets</th>\n",
       "      <th>TotalBackwardPackets</th>\n",
       "      <th>TotalLengthofFwdPackets</th>\n",
       "      <th>TotalLengthofBwdPackets</th>\n",
       "      <th>FwdPacketLengthMax</th>\n",
       "      <th>FwdPacketLengthMin</th>\n",
       "      <th>FwdPacketLengthMean</th>\n",
       "      <th>FwdPacketLengthStd</th>\n",
       "      <th>...</th>\n",
       "      <th>min_seg_size_forward</th>\n",
       "      <th>ActiveMean</th>\n",
       "      <th>ActiveStd</th>\n",
       "      <th>ActiveMax</th>\n",
       "      <th>ActiveMin</th>\n",
       "      <th>IdleMean</th>\n",
       "      <th>IdleStd</th>\n",
       "      <th>IdleMax</th>\n",
       "      <th>IdleMin</th>\n",
       "      <th>Label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>54865.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>55054.0</td>\n",
       "      <td>109.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>55055.0</td>\n",
       "      <td>52.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>46236.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>54863.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>BENIGN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 79 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   DestinationPort  FlowDuration  TotalFwdPackets  TotalBackwardPackets  \\\n",
       "0          54865.0           3.0              2.0                   0.0   \n",
       "1          55054.0         109.0              1.0                   1.0   \n",
       "2          55055.0          52.0              1.0                   1.0   \n",
       "3          46236.0          34.0              1.0                   1.0   \n",
       "4          54863.0           3.0              2.0                   0.0   \n",
       "\n",
       "   TotalLengthofFwdPackets  TotalLengthofBwdPackets  FwdPacketLengthMax  \\\n",
       "0                     12.0                      0.0                 6.0   \n",
       "1                      6.0                      6.0                 6.0   \n",
       "2                      6.0                      6.0                 6.0   \n",
       "3                      6.0                      6.0                 6.0   \n",
       "4                     12.0                      0.0                 6.0   \n",
       "\n",
       "   FwdPacketLengthMin  FwdPacketLengthMean  FwdPacketLengthStd  ...  \\\n",
       "0                 6.0                  6.0                 0.0  ...   \n",
       "1                 6.0                  6.0                 0.0  ...   \n",
       "2                 6.0                  6.0                 0.0  ...   \n",
       "3                 6.0                  6.0                 0.0  ...   \n",
       "4                 6.0                  6.0                 0.0  ...   \n",
       "\n",
       "   min_seg_size_forward  ActiveMean  ActiveStd  ActiveMax  ActiveMin  \\\n",
       "0                  20.0         0.0        0.0        0.0        0.0   \n",
       "1                  20.0         0.0        0.0        0.0        0.0   \n",
       "2                  20.0         0.0        0.0        0.0        0.0   \n",
       "3                  20.0         0.0        0.0        0.0        0.0   \n",
       "4                  20.0         0.0        0.0        0.0        0.0   \n",
       "\n",
       "   IdleMean  IdleStd  IdleMax  IdleMin   Label  \n",
       "0       0.0      0.0      0.0      0.0  BENIGN  \n",
       "1       0.0      0.0      0.0      0.0  BENIGN  \n",
       "2       0.0      0.0      0.0      0.0  BENIGN  \n",
       "3       0.0      0.0      0.0      0.0  BENIGN  \n",
       "4       0.0      0.0      0.0      0.0  BENIGN  \n",
       "\n",
       "[5 rows x 79 columns]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# Saving cleaned dataset.\n",
    "\n",
    "#dataset.to_csv(\"Dataset_clean_dropna.csv\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# Data visualization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "So, by now we know our dataset has 78 features and is split into 15 categories (14 attacks and 1 \"normal\" state).\n",
    "Next step is to try and visualize what the dataset looks like in feature space.\n",
    "For this we will use principal component analysis (PCA) to reduce dimensionality and then pass the reduced dataset to t-SNE (t - Distributed Stohastic Neighbor Entities) for visual representation in 2D space."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "from sklearn.manifold import TSNE\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# We are going to pick 10.000 random rows from the dataset for visualization purposes.\n",
    "# Setting the random seed for reproducability of results.\n",
    "\n",
    "np.random.seed(42)\n",
    "\n",
    "rand_perm = np.random.permutation(dataset.shape[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 214,
   "metadata": {
    "collapsed": true,
    "hidden": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\marko\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\pandas\\core\\indexing.py:1418: FutureWarning: \n",
      "Passing list-likes to .loc or [] with any missing label will raise\n",
      "KeyError in the future, you can use .reindex() as an alternative.\n",
      "\n",
      "See the documentation here:\n",
      "https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#deprecate-loc-reindex-listlike\n",
      "  return self._getitem_tuple(key)\n"
     ]
    }
   ],
   "source": [
    "feature_cols = dataset.columns[:-1]\n",
    "\n",
    "dataset_subset = dataset.loc[rand_perm[:10000],:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 215,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "dataset_subset = dataset_subset.replace([np.inf, -np.inf], np.nan)\n",
    "dataset_subset.dropna(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 216,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "data_subset = dataset_subset[feature_cols].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 195,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9999379655912742"
      ]
     },
     "execution_count": 195,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Performing the principal component analysis. With just 19 components the variance ratio remains 99%, which is great.\n",
    "\n",
    "pca = PCA(n_components=19)\n",
    "pca_res = pca.fit_transform(data_subset)\n",
    "\n",
    "data_subset = None\n",
    "np.sum(pca.explained_variance_ratio_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 217,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "done\n"
     ]
    }
   ],
   "source": [
    "# Computing t-SNE.\n",
    "\n",
    "tsne = TSNE(n_components=2, verbose=0, perplexity=40, n_iter=1000)\n",
    "tsne_res = tsne.fit_transform(data_subset)\n",
    "print(\"done\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 218,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "dataset_subset['tsne_firstD'] = tsne_res[:,0]\n",
    "dataset_subset['tsne_secondD'] = tsne_res[:,1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "metadata": {
    "hidden": true,
    "hide_input": false,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1eb8e603080>"
      ]
     },
     "execution_count": 220,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x1152 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(16,16))\n",
    "\n",
    "sns.scatterplot(\n",
    "    x=\"tsne_firstD\", y=\"tsne_secondD\",\n",
    "    palette=sns.color_palette(\"hls\", colors),\n",
    "    data=dataset_subset,\n",
    "    hue=\"Label\",\n",
    "    legend=\"full\",\n",
    "    alpha=0.3\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "From the cell above we can see distribution of the data in 2D space. It is obvious that attacks are not spatialy well separated from normal state. Clusters of attacks can hardly be seen, instead they are found in the same place as the \"normal state\" datatpoints.\n",
    "\n",
    "This insight leads us to conclude that the ML model will probably have some issues with this kind of data. ML model will have to be chosen with this in mind."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# Data preparation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "In this chapter, final data preparation steps are taken before we use the data for model traning and testing.\n",
    "\n",
    "These steps include:\n",
    "\n",
    "* Data scaling\n",
    "* Label encoding\n",
    "* Data splitting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "#dataset = pd.read_csv(\"Dataset_clean_dropna.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "## Scaling the data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "The next few cells contain the code for scaling the data into the size adequate for the ML algorithm. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# Splitting dataset into features and labels.\n",
    "\n",
    "labels = dataset['Label']\n",
    "features = dataset.loc[:, dataset.columns != 'Label'].astype('float64')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "hidden": true
   },
   "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>DestinationPort</th>\n",
       "      <th>FlowDuration</th>\n",
       "      <th>TotalFwdPackets</th>\n",
       "      <th>TotalBackwardPackets</th>\n",
       "      <th>TotalLengthofFwdPackets</th>\n",
       "      <th>TotalLengthofBwdPackets</th>\n",
       "      <th>FwdPacketLengthMax</th>\n",
       "      <th>FwdPacketLengthMin</th>\n",
       "      <th>FwdPacketLengthMean</th>\n",
       "      <th>FwdPacketLengthStd</th>\n",
       "      <th>...</th>\n",
       "      <th>act_data_pkt_fwd</th>\n",
       "      <th>min_seg_size_forward</th>\n",
       "      <th>ActiveMean</th>\n",
       "      <th>ActiveStd</th>\n",
       "      <th>ActiveMax</th>\n",
       "      <th>ActiveMin</th>\n",
       "      <th>IdleMean</th>\n",
       "      <th>IdleStd</th>\n",
       "      <th>IdleMax</th>\n",
       "      <th>IdleMin</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>54865.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>55054.0</td>\n",
       "      <td>109.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>55055.0</td>\n",
       "      <td>52.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>46236.0</td>\n",
       "      <td>34.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>54863.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 78 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   DestinationPort  FlowDuration  TotalFwdPackets  TotalBackwardPackets  \\\n",
       "0          54865.0           3.0              2.0                   0.0   \n",
       "1          55054.0         109.0              1.0                   1.0   \n",
       "2          55055.0          52.0              1.0                   1.0   \n",
       "3          46236.0          34.0              1.0                   1.0   \n",
       "4          54863.0           3.0              2.0                   0.0   \n",
       "\n",
       "   TotalLengthofFwdPackets  TotalLengthofBwdPackets  FwdPacketLengthMax  \\\n",
       "0                     12.0                      0.0                 6.0   \n",
       "1                      6.0                      6.0                 6.0   \n",
       "2                      6.0                      6.0                 6.0   \n",
       "3                      6.0                      6.0                 6.0   \n",
       "4                     12.0                      0.0                 6.0   \n",
       "\n",
       "   FwdPacketLengthMin  FwdPacketLengthMean  FwdPacketLengthStd  ...  \\\n",
       "0                 6.0                  6.0                 0.0  ...   \n",
       "1                 6.0                  6.0                 0.0  ...   \n",
       "2                 6.0                  6.0                 0.0  ...   \n",
       "3                 6.0                  6.0                 0.0  ...   \n",
       "4                 6.0                  6.0                 0.0  ...   \n",
       "\n",
       "   act_data_pkt_fwd  min_seg_size_forward  ActiveMean  ActiveStd  ActiveMax  \\\n",
       "0               1.0                  20.0         0.0        0.0        0.0   \n",
       "1               0.0                  20.0         0.0        0.0        0.0   \n",
       "2               0.0                  20.0         0.0        0.0        0.0   \n",
       "3               0.0                  20.0         0.0        0.0        0.0   \n",
       "4               1.0                  20.0         0.0        0.0        0.0   \n",
       "\n",
       "   ActiveMin  IdleMean  IdleStd  IdleMax  IdleMin  \n",
       "0        0.0       0.0      0.0      0.0      0.0  \n",
       "1        0.0       0.0      0.0      0.0      0.0  \n",
       "2        0.0       0.0      0.0      0.0      0.0  \n",
       "3        0.0       0.0      0.0      0.0      0.0  \n",
       "4        0.0       0.0      0.0      0.0      0.0  \n",
       "\n",
       "[5 rows x 78 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# For scaling the data, we use RobustScaler class from sklearn.\n",
    "\n",
    "from sklearn.preprocessing import MinMaxScaler, StandardScaler, RobustScaler"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "For scaling the data we used RobustScaler class from sklearn. RobustScaler is used to perserve outliers in the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "hidden": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "scaler = RobustScaler()\n",
    "scaler.fit(features)\n",
    "\n",
    "features = scaler.transform(features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1.40474359e+02, -1.08842743e-02,  0.00000000e+00, -5.00000000e-01,\n",
       "       -1.75000000e-01, -1.14326041e-01, -1.66666667e-01,  1.08108108e-01,\n",
       "       -6.59574468e-01,  0.00000000e+00, -1.17233294e-01,  0.00000000e+00,\n",
       "       -4.06250000e-01,  0.00000000e+00,  5.17015897e+01,  4.18340621e+01,\n",
       "       -3.64260699e-02, -8.53215162e-03, -9.45038074e-03, -1.49253731e-02,\n",
       "       -8.85915295e-06, -5.67715025e-05,  0.00000000e+00, -9.42805378e-06,\n",
       "        0.00000000e+00, -1.85178327e-05, -9.49938254e-05,  0.00000000e+00,\n",
       "       -2.26719240e-05, -6.38297872e-02,  0.00000000e+00,  0.00000000e+00,\n",
       "        0.00000000e+00,  0.00000000e+00, -2.14285714e-01, -3.57142857e-01,\n",
       "        7.70059825e+01, -5.27501871e-02,  1.08108108e-01, -7.23589001e-02,\n",
       "       -3.73417722e-01, -9.42324454e-02, -8.73549112e-03,  0.00000000e+00,\n",
       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  1.00000000e+00,\n",
       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00, -1.00000000e+00,\n",
       "       -4.14634146e-01, -6.59574468e-01, -4.06250000e-01, -2.14285714e-01,\n",
       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00, -1.75000000e-01,\n",
       "       -5.00000000e-01, -1.14326041e-01, -2.66080801e-02, -4.23728814e-03,\n",
       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
       "        0.00000000e+00,  0.00000000e+00])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking if scaling has been succesful.\n",
    "features[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "## Label encoding"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Label encoding is done when dataset contains categorical values (ex. 0-5, A/B/C, 55+). It is used to turn categorical values into numerical values by replacing data categories with integers starting with 0."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "hidden": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\users\\marko\\appdata\\local\\programs\\python\\python36\\lib\\site-packages\\numpy\\lib\\arraysetops.py:568: FutureWarning: elementwise comparison failed; returning scalar instead, but in the future will perform elementwise comparison\n",
      "  mask |= (ar1 == a)\n"
     ]
    }
   ],
   "source": [
    "# No need to do previous operations, just load clean saved dataset.\n",
    "\n",
    "#dataset = pd.read_csv('Dataset_clean.csv', index_col=[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import LabelEncoder"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "'Lables' column contains categorical values - 15 of them (14 types of attacks in our dataset +  1 normal state).\n",
    "\n",
    "To convert this into numerical values we will use 'LabelEncoder' class from sklearn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "LE = LabelEncoder()\n",
    "\n",
    "LE.fit(labels)\n",
    "labels = LE.transform(labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0,  1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14])"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Labels have been replaced with integers.\n",
    "\n",
    "np.unique(labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['BENIGN', 'DDoS', 'PortScan', 'Bot', 'Infiltration',\n",
       "       'Web_Attack_Brute_Force', 'Web_Attack_XSS',\n",
       "       'Web_Attack_Sql_Injection', 'FTPPatator', 'SSHPatator',\n",
       "       'DoS_slowloris', 'DoS_Slowhttptest', 'DoS_Hulk', 'DoS_GoldenEye',\n",
       "       'Heartbleed'], dtype=object)"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking that encoding reversal works.\n",
    "\n",
    "d = LE.inverse_transform(labels)\n",
    "d = pd.Series(d)\n",
    "d.unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true,
    "hidden": true
   },
   "source": [
    "## Splitting the data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Final step to data preparation is splitting the data into traning and testing sets. For this there already exists _sklearn_ function that does all the splitting for us. This step is important so we can have representative data for evaluating our model. Both train and test samples should contain similar data variance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "hidden": true,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1960425, 78), (490107, 78), (1960425,), (490107,))"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# The next step is to split training and testing data. For this we will use sklearn function train_test_split().\n",
    "\n",
    "features_train, features_test, labels_train, labels_test = train_test_split(features, labels, test_size=.2)\n",
    "\n",
    "features_train.shape, features_test.shape, labels_train.shape, labels_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# Clearing variables.\n",
    "\n",
    "dataset = None\n",
    "finite = None\n",
    "labl = None\n",
    "d = None\n",
    "features = None\n",
    "labels = None"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# Model training"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "For completing this task we chose to use a neural network. Specifically, the multi-layer perceptron, more specifically, feedforward neural network multi-class classifier with backpropagating algorithm. NN will be used to classifiy 14 different attacks and 1 normal state, as we saw from the labels in previous chapters.\n",
    "\n",
    "In this chapter we go by explaning parts of the network and its hyperparameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import tensorflow as tf\n",
    "import datetime\n",
    "\n",
    "#%load_ext tensorboard"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "Our tensorflow Sequential model has 3 layers. Input, 1 hidden and an output layer.\n",
    "\n",
    "* Input layer has 78 neurons, one for each feature.\n",
    "* Hidden layer has 67 neurons, this number has been calculated by [formula](https://www.heatonresearch.com/2017/06/01/hidden-layers.html) 2/3 the number of input neurons + number of output neurons.\n",
    "* Output layer has 15 neurons, one for each class we predict.\n",
    "\n",
    "For activation functions, we used standard functions for multi-class classification tasks - ReLu for hidden layer and _softmax_ function for output layer.\n",
    "\n",
    "Finally, we use Dropout parameter set to 0.2 for randomly shutting off 20% of neurons in each learning iteration. This technique is used for decreasing overfitting thereby incresing network accuracy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "model = tf.keras.models.Sequential([\n",
    "    \n",
    "    tf.keras.layers.Flatten(input_shape=(78,)),\n",
    "    tf.keras.layers.Dense(67, activation='relu'),\n",
    "    tf.keras.layers.Dropout(0.2),\n",
    "    tf.keras.layers.Dense(15, activation='softmax')\n",
    "])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "For learning rate optimization we used Adam optimizer.\n",
    "Loss function used is sparse categorical crossentropy, which is standard for multiclass classification problems."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "model.compile(optimizer='adam',\n",
    "             loss='sparse_categorical_crossentropy',\n",
    "             metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "import os"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "In the next cell we setup training logs for tensorboard as well as some tensorboard callbacks.\n",
    "\n",
    "* tensorboard - callback that logs training data.\n",
    "* EarlyStopping - callback that monitors 'loss (function)' metric and if the loss function does not get better in tne hext 10 iterations, callback stops the training and resotres the network with best weights up untill that iteration."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "log_dir = os.path.join(\n",
    "    \"train_logs\",\n",
    "    datetime.datetime.now().strftime(\"%Y%m%d-%H%M%S\"),\n",
    ")\n",
    "\n",
    "# TF callback that sets up TensorBoard with training logs.\n",
    "tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir=log_dir, histogram_freq=1)\n",
    "\n",
    "# TF callback that stops training when best value of validationi loss function is reached. It also\n",
    "# restores weights from the best training iteration.\n",
    "eary_stop_callback = tf.keras.callbacks.EarlyStopping(monitor='loss', patience=10, restore_best_weights=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 1960425 samples\n",
      "Epoch 1/100\n",
      "1960425/1960425 [==============================] - 66s 33us/sample - loss: 1.3201 - accuracy: 0.9105\n",
      "Epoch 2/100\n",
      "1960425/1960425 [==============================] - 63s 32us/sample - loss: 1.2584 - accuracy: 0.9104\n",
      "Epoch 3/100\n",
      "1960425/1960425 [==============================] - 63s 32us/sample - loss: 1.2423 - accuracy: 0.9105\n",
      "Epoch 4/100\n",
      "1960425/1960425 [==============================] - 62s 32us/sample - loss: 1.9466 - accuracy: 0.9105\n",
      "Epoch 5/100\n",
      "1960425/1960425 [==============================] - 62s 32us/sample - loss: 1.1178 - accuracy: 0.9104\n",
      "Epoch 6/100\n",
      "1960425/1960425 [==============================] - 60s 31us/sample - loss: 1.2169 - accuracy: 0.9105\n",
      "Epoch 7/100\n",
      "1960425/1960425 [==============================] - 62s 32us/sample - loss: 0.8587 - accuracy: 0.9104\n",
      "Epoch 8/100\n",
      "1960425/1960425 [==============================] - ETA: 0s - loss: 0.7722 - accuracy: 0.91 - 63s 32us/sample - loss: 0.7720 - accuracy: 0.9104\n",
      "Epoch 9/100\n",
      "1960425/1960425 [==============================] - 62s 32us/sample - loss: 1.1078 - accuracy: 0.9104\n",
      "Epoch 10/100\n",
      "1960425/1960425 [==============================] - 60s 30us/sample - loss: 1.1971 - accuracy: 0.9104\n",
      "Epoch 11/100\n",
      "1960425/1960425 [==============================] - 60s 30us/sample - loss: 1.3754 - accuracy: 0.9105\n",
      "Epoch 12/100\n",
      "1960425/1960425 [==============================] - 59s 30us/sample - loss: 1.9107 - accuracy: 0.9104\n",
      "Epoch 13/100\n",
      "1960425/1960425 [==============================] - 59s 30us/sample - loss: 0.9216 - accuracy: 0.9105\n",
      "Epoch 14/100\n",
      "1960425/1960425 [==============================] - 63s 32us/sample - loss: 0.9553 - accuracy: 0.9105\n",
      "Epoch 15/100\n",
      "1960425/1960425 [==============================] - 60s 31us/sample - loss: 1.0729 - accuracy: 0.9105\n",
      "Epoch 16/100\n",
      "1960425/1960425 [==============================] - 59s 30us/sample - loss: 1.8300 - accuracy: 0.9105\n",
      "Epoch 17/100\n",
      "1960425/1960425 [==============================] - 59s 30us/sample - loss: 1.0752 - accuracy: 0.9105\n",
      "Epoch 18/100\n",
      "1960425/1960425 [==============================] - 59s 30us/sample - loss: 1.1702 - accuracy: 0.9105\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.keras.callbacks.History at 0x254f972de10>"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(features_train,\n",
    "          labels_train,\n",
    "          epochs=100,\n",
    "          callbacks=[tensorboard_callback, eary_stop_callback])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "We can see that training stoped after 18 out of 100 epochs due to 'loss' function metric not changing much in the previous 10 epochs.\n",
    "\n",
    "After training we evaluate model accuracy (next cell), and find that our model predicts attacks with **91.2% accuracy**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "490107/1 - 8s - loss: 0.2958 - accuracy: 0.9124\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.28660620927896285, 0.9124497]"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Evaluating model accuracy.\n",
    "model.evaluate(features_test, labels_test, verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# Saving the model.\n",
    "\n",
    "model.save('saved_models/IDS_model_' + datetime.datetime.now().strftime(\"%Y%m%d-%H%M%S\") + '.h5')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "# Conclusion"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "hidden": true
   },
   "source": [
    "In this project we made a neural network classifier that can predict 14 network/web attacks and normal traffic with 91% accuracy. This model is proof-of-concept that feedforward neural network with backpropagation algorithm can be used for classifying attacks in anomaly-based intrusion detection systems.\n",
    "\n",
    "\n",
    "**Propositions**\n",
    "\n",
    "We propose a couple of solutions for improving model accuracy as well as use of some other neural network architectures.\n",
    "\n",
    "Accuracy of this model can probably be improved by _feature engineering_ and _feature selection_. Picking the features that have the most influence on the model.\n",
    "\n",
    "Regarding this model, we propose tuning the model hyperparameters. Changing the hidden layer activation function, early stopping callback, dropout, optimizer and loss function should increase accuracy by some extent. Another way, albeit more complicated and resource intense is to use a genetic algorithm to evolve the best neural network arhitecture for this specific task.\n",
    "\n",
    "Finally, we propose the usage of some other ML algorithms. Random forest classifiers have been used in intrusion detection system for a while now. Alternatively, we found some sources using autoencoders for anomaly detection."
   ]
  }
 ],
 "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": "3.6.8"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "375.994px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "position": {
    "height": "485.188px",
    "left": "645.125px",
    "right": "20px",
    "top": "174.938px",
    "width": "628.75px"
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}