windows Pe malware detection using Ensamble learning.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "b1c981f6",
   "metadata": {},
   "source": [
    "# Cyber attacks analysis methods Course - Final Project"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b80f535",
   "metadata": {},
   "source": [
    "### Our research was based on the above article:\n",
    "[Windows Pe malware detection using Ensamble learning](https://www.mdpi.com/2227-9709/8/1/10) \n",
    "\n",
    "### With the dataset from:\n",
    "[Kaggle](https://www.kaggle.com/datasets/amauricio/pe-files-malwares?resource=download)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7847c216",
   "metadata": {},
   "source": [
    "# Windows Pe malware detection using Ensamble learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "f63125bc",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3d39b029",
   "metadata": {},
   "source": [
    "## Data Collection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e4e50174",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The data contains 19611 rows, and 79 columns\n"
     ]
    },
    {
     "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>Name</th>\n",
       "      <th>e_magic</th>\n",
       "      <th>e_cblp</th>\n",
       "      <th>e_cp</th>\n",
       "      <th>e_crlc</th>\n",
       "      <th>e_cparhdr</th>\n",
       "      <th>e_minalloc</th>\n",
       "      <th>e_maxalloc</th>\n",
       "      <th>e_ss</th>\n",
       "      <th>e_sp</th>\n",
       "      <th>...</th>\n",
       "      <th>SectionMaxChar</th>\n",
       "      <th>SectionMainChar</th>\n",
       "      <th>DirectoryEntryImport</th>\n",
       "      <th>DirectoryEntryImportSize</th>\n",
       "      <th>DirectoryEntryExport</th>\n",
       "      <th>ImageDirectoryEntryExport</th>\n",
       "      <th>ImageDirectoryEntryImport</th>\n",
       "      <th>ImageDirectoryEntryResource</th>\n",
       "      <th>ImageDirectoryEntryException</th>\n",
       "      <th>ImageDirectoryEntrySecurity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>VirusShare_a878ba26000edaac5c98eff4432723b3</td>\n",
       "      <td>23117</td>\n",
       "      <td>144</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>65535</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>...</td>\n",
       "      <td>3758096608</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>152</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>54440</td>\n",
       "      <td>77824</td>\n",
       "      <td>73728</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>VirusShare_ef9130570fddc174b312b2047f5f4cf0</td>\n",
       "      <td>23117</td>\n",
       "      <td>144</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>65535</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>...</td>\n",
       "      <td>3791650880</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>311</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>262276</td>\n",
       "      <td>294912</td>\n",
       "      <td>0</td>\n",
       "      <td>346112</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>VirusShare_ef84cdeba22be72a69b198213dada81a</td>\n",
       "      <td>23117</td>\n",
       "      <td>144</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>65535</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>...</td>\n",
       "      <td>3221225536</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>176</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>36864</td>\n",
       "      <td>40960</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>VirusShare_6bf3608e60ebc16cbcff6ed5467d469e</td>\n",
       "      <td>23117</td>\n",
       "      <td>144</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>65535</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>...</td>\n",
       "      <td>3224371328</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>155</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>356352</td>\n",
       "      <td>1003520</td>\n",
       "      <td>0</td>\n",
       "      <td>14109472</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>VirusShare_2cc94d952b2efb13c7d6bbe0dd59d3fb</td>\n",
       "      <td>23117</td>\n",
       "      <td>144</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>65535</td>\n",
       "      <td>0</td>\n",
       "      <td>184</td>\n",
       "      <td>...</td>\n",
       "      <td>3227516992</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>43</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>61440</td>\n",
       "      <td>73728</td>\n",
       "      <td>0</td>\n",
       "      <td>90624</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 79 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                          Name  e_magic  e_cblp  e_cp  e_crlc  \\\n",
       "0  VirusShare_a878ba26000edaac5c98eff4432723b3    23117     144     3       0   \n",
       "1  VirusShare_ef9130570fddc174b312b2047f5f4cf0    23117     144     3       0   \n",
       "2  VirusShare_ef84cdeba22be72a69b198213dada81a    23117     144     3       0   \n",
       "3  VirusShare_6bf3608e60ebc16cbcff6ed5467d469e    23117     144     3       0   \n",
       "4  VirusShare_2cc94d952b2efb13c7d6bbe0dd59d3fb    23117     144     3       0   \n",
       "\n",
       "   e_cparhdr  e_minalloc  e_maxalloc  e_ss  e_sp  ...  SectionMaxChar  \\\n",
       "0          4           0       65535     0   184  ...      3758096608   \n",
       "1          4           0       65535     0   184  ...      3791650880   \n",
       "2          4           0       65535     0   184  ...      3221225536   \n",
       "3          4           0       65535     0   184  ...      3224371328   \n",
       "4          4           0       65535     0   184  ...      3227516992   \n",
       "\n",
       "   SectionMainChar  DirectoryEntryImport  DirectoryEntryImportSize  \\\n",
       "0                0                     7                       152   \n",
       "1                0                    16                       311   \n",
       "2                0                     6                       176   \n",
       "3                0                     8                       155   \n",
       "4                0                     2                        43   \n",
       "\n",
       "   DirectoryEntryExport  ImageDirectoryEntryExport  ImageDirectoryEntryImport  \\\n",
       "0                     0                          0                      54440   \n",
       "1                     0                          0                     262276   \n",
       "2                     0                          0                      36864   \n",
       "3                     0                          0                     356352   \n",
       "4                     0                          0                      61440   \n",
       "\n",
       "   ImageDirectoryEntryResource  ImageDirectoryEntryException  \\\n",
       "0                        77824                         73728   \n",
       "1                       294912                             0   \n",
       "2                        40960                             0   \n",
       "3                      1003520                             0   \n",
       "4                        73728                             0   \n",
       "\n",
       "   ImageDirectoryEntrySecurity  \n",
       "0                            0  \n",
       "1                       346112  \n",
       "2                            0  \n",
       "3                     14109472  \n",
       "4                        90624  \n",
       "\n",
       "[5 rows x 79 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"dataset_malwares.csv\")\n",
    "print(\"The data contains\", df.shape[0],\"rows, and\",df.shape[1],\"columns\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "64639e6b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 19611 entries, 0 to 19610\n",
      "Data columns (total 79 columns):\n",
      " #   Column                        Non-Null Count  Dtype  \n",
      "---  ------                        --------------  -----  \n",
      " 0   Name                          19611 non-null  object \n",
      " 1   e_magic                       19611 non-null  int64  \n",
      " 2   e_cblp                        19611 non-null  int64  \n",
      " 3   e_cp                          19611 non-null  int64  \n",
      " 4   e_crlc                        19611 non-null  int64  \n",
      " 5   e_cparhdr                     19611 non-null  int64  \n",
      " 6   e_minalloc                    19611 non-null  int64  \n",
      " 7   e_maxalloc                    19611 non-null  int64  \n",
      " 8   e_ss                          19611 non-null  int64  \n",
      " 9   e_sp                          19611 non-null  int64  \n",
      " 10  e_csum                        19611 non-null  int64  \n",
      " 11  e_ip                          19611 non-null  int64  \n",
      " 12  e_cs                          19611 non-null  int64  \n",
      " 13  e_lfarlc                      19611 non-null  int64  \n",
      " 14  e_ovno                        19611 non-null  int64  \n",
      " 15  e_oemid                       19611 non-null  int64  \n",
      " 16  e_oeminfo                     19611 non-null  int64  \n",
      " 17  e_lfanew                      19611 non-null  int64  \n",
      " 18  Machine                       19611 non-null  int64  \n",
      " 19  NumberOfSections              19611 non-null  int64  \n",
      " 20  TimeDateStamp                 19611 non-null  int64  \n",
      " 21  PointerToSymbolTable          19611 non-null  int64  \n",
      " 22  NumberOfSymbols               19611 non-null  int64  \n",
      " 23  SizeOfOptionalHeader          19611 non-null  int64  \n",
      " 24  Characteristics               19611 non-null  int64  \n",
      " 25  Magic                         19611 non-null  int64  \n",
      " 26  MajorLinkerVersion            19611 non-null  int64  \n",
      " 27  MinorLinkerVersion            19611 non-null  int64  \n",
      " 28  SizeOfCode                    19611 non-null  int64  \n",
      " 29  SizeOfInitializedData         19611 non-null  int64  \n",
      " 30  SizeOfUninitializedData       19611 non-null  int64  \n",
      " 31  AddressOfEntryPoint           19611 non-null  int64  \n",
      " 32  BaseOfCode                    19611 non-null  int64  \n",
      " 33  ImageBase                     19611 non-null  int64  \n",
      " 34  SectionAlignment              19611 non-null  int64  \n",
      " 35  FileAlignment                 19611 non-null  int64  \n",
      " 36  MajorOperatingSystemVersion   19611 non-null  int64  \n",
      " 37  MinorOperatingSystemVersion   19611 non-null  int64  \n",
      " 38  MajorImageVersion             19611 non-null  int64  \n",
      " 39  MinorImageVersion             19611 non-null  int64  \n",
      " 40  MajorSubsystemVersion         19611 non-null  int64  \n",
      " 41  MinorSubsystemVersion         19611 non-null  int64  \n",
      " 42  SizeOfHeaders                 19611 non-null  int64  \n",
      " 43  CheckSum                      19611 non-null  int64  \n",
      " 44  SizeOfImage                   19611 non-null  int64  \n",
      " 45  Subsystem                     19611 non-null  int64  \n",
      " 46  DllCharacteristics            19611 non-null  int64  \n",
      " 47  SizeOfStackReserve            19611 non-null  int64  \n",
      " 48  SizeOfStackCommit             19611 non-null  int64  \n",
      " 49  SizeOfHeapReserve             19611 non-null  int64  \n",
      " 50  SizeOfHeapCommit              19611 non-null  int64  \n",
      " 51  LoaderFlags                   19611 non-null  int64  \n",
      " 52  NumberOfRvaAndSizes           19611 non-null  int64  \n",
      " 53  Malware                       19611 non-null  int64  \n",
      " 54  SuspiciousImportFunctions     19611 non-null  int64  \n",
      " 55  SuspiciousNameSection         19611 non-null  int64  \n",
      " 56  SectionsLength                19611 non-null  int64  \n",
      " 57  SectionMinEntropy             19611 non-null  float64\n",
      " 58  SectionMaxEntropy             19611 non-null  int64  \n",
      " 59  SectionMinRawsize             19611 non-null  int64  \n",
      " 60  SectionMaxRawsize             19611 non-null  int64  \n",
      " 61  SectionMinVirtualsize         19611 non-null  int64  \n",
      " 62  SectionMaxVirtualsize         19611 non-null  int64  \n",
      " 63  SectionMaxPhysical            19611 non-null  int64  \n",
      " 64  SectionMinPhysical            19611 non-null  int64  \n",
      " 65  SectionMaxVirtual             19611 non-null  int64  \n",
      " 66  SectionMinVirtual             19611 non-null  int64  \n",
      " 67  SectionMaxPointerData         19611 non-null  int64  \n",
      " 68  SectionMinPointerData         19611 non-null  int64  \n",
      " 69  SectionMaxChar                19611 non-null  int64  \n",
      " 70  SectionMainChar               19611 non-null  int64  \n",
      " 71  DirectoryEntryImport          19611 non-null  int64  \n",
      " 72  DirectoryEntryImportSize      19611 non-null  int64  \n",
      " 73  DirectoryEntryExport          19611 non-null  int64  \n",
      " 74  ImageDirectoryEntryExport     19611 non-null  int64  \n",
      " 75  ImageDirectoryEntryImport     19611 non-null  int64  \n",
      " 76  ImageDirectoryEntryResource   19611 non-null  int64  \n",
      " 77  ImageDirectoryEntryException  19611 non-null  int64  \n",
      " 78  ImageDirectoryEntrySecurity   19611 non-null  int64  \n",
      "dtypes: float64(1), int64(77), object(1)\n",
      "memory usage: 11.8+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "6fd47e6b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.Malware.value_counts().plot(kind='bar')\n",
    "plt.xlabel('Label')\n",
    "plt.ylabel('samples')\n",
    "plt.title('Distribution of samples in df')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "4ea9dde3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1600x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Assuming df is your DataFrame with 79 columns\n",
    "correlation_with_malware = df.corr()['Malware'].drop('Malware')\n",
    "\n",
    "# Plotting the correlation values\n",
    "plt.figure(figsize=(16, 8))\n",
    "correlation_with_malware.plot(kind='bar', color='skyblue')\n",
    "plt.title('Correlation of Features with \"Malware\"')\n",
    "plt.xlabel('Features')\n",
    "plt.ylabel('Correlation')\n",
    "plt.xticks(rotation=90)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "1f0e96eb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top 5 Features with Highest Correlation to 'Malware':\n",
      "SectionMaxChar               0.393282\n",
      "SizeOfStackReserve           0.251791\n",
      "SuspiciousImportFunctions    0.216656\n",
      "DllCharacteristics           0.197023\n",
      "e_maxalloc                   0.186079\n",
      "Name: Malware, dtype: float64 \n",
      "\n",
      "Top 5 Features with Lowest Correlation to 'Malware':\n",
      "MajorSubsystemVersion   -0.611621\n",
      "Subsystem               -0.492813\n",
      "Magic                   -0.287414\n",
      "Machine                 -0.287413\n",
      "SizeOfOptionalHeader    -0.255692\n",
      "Name: Malware, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Find the top 5 correlations\n",
    "ten_highest_corr = correlation_with_malware.nlargest(5)\n",
    "ten_lowest_corr = correlation_with_malware.nsmallest(5)\n",
    "\n",
    "# Print the results\n",
    "print(\"Top 5 Features with Highest Correlation to 'Malware':\")\n",
    "print(ten_highest_corr,\"\\n\")\n",
    "\n",
    "print(\"Top 5 Features with Lowest Correlation to 'Malware':\")\n",
    "print(ten_lowest_corr)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "21b292fb",
   "metadata": {},
   "source": [
    "## Dimensionality Reduction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "3ed8029d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(19611, 56)\n"
     ]
    },
    {
     "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>PC1</th>\n",
       "      <th>PC2</th>\n",
       "      <th>PC3</th>\n",
       "      <th>PC4</th>\n",
       "      <th>PC5</th>\n",
       "      <th>PC6</th>\n",
       "      <th>PC7</th>\n",
       "      <th>PC8</th>\n",
       "      <th>PC9</th>\n",
       "      <th>PC10</th>\n",
       "      <th>...</th>\n",
       "      <th>PC47</th>\n",
       "      <th>PC48</th>\n",
       "      <th>PC49</th>\n",
       "      <th>PC50</th>\n",
       "      <th>PC51</th>\n",
       "      <th>PC52</th>\n",
       "      <th>PC53</th>\n",
       "      <th>PC54</th>\n",
       "      <th>PC55</th>\n",
       "      <th>Malware</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.121848</td>\n",
       "      <td>-0.109631</td>\n",
       "      <td>0.677826</td>\n",
       "      <td>1.666928</td>\n",
       "      <td>1.618763</td>\n",
       "      <td>-0.550915</td>\n",
       "      <td>0.724746</td>\n",
       "      <td>-0.201950</td>\n",
       "      <td>-0.829159</td>\n",
       "      <td>-0.149451</td>\n",
       "      <td>...</td>\n",
       "      <td>0.065499</td>\n",
       "      <td>0.062655</td>\n",
       "      <td>-0.133934</td>\n",
       "      <td>0.168381</td>\n",
       "      <td>-0.588060</td>\n",
       "      <td>0.035701</td>\n",
       "      <td>0.005702</td>\n",
       "      <td>0.009438</td>\n",
       "      <td>0.073707</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.294837</td>\n",
       "      <td>-0.135038</td>\n",
       "      <td>0.612546</td>\n",
       "      <td>-0.003972</td>\n",
       "      <td>0.179251</td>\n",
       "      <td>0.200744</td>\n",
       "      <td>0.551912</td>\n",
       "      <td>0.277117</td>\n",
       "      <td>0.068057</td>\n",
       "      <td>-1.479715</td>\n",
       "      <td>...</td>\n",
       "      <td>0.201665</td>\n",
       "      <td>-0.068814</td>\n",
       "      <td>-0.228309</td>\n",
       "      <td>0.154042</td>\n",
       "      <td>-0.806254</td>\n",
       "      <td>-0.028854</td>\n",
       "      <td>-0.015292</td>\n",
       "      <td>0.066456</td>\n",
       "      <td>0.122343</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.296454</td>\n",
       "      <td>-0.122306</td>\n",
       "      <td>0.340623</td>\n",
       "      <td>-0.655986</td>\n",
       "      <td>-1.082754</td>\n",
       "      <td>0.059316</td>\n",
       "      <td>-0.495534</td>\n",
       "      <td>-0.429369</td>\n",
       "      <td>0.711304</td>\n",
       "      <td>-0.653683</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.023068</td>\n",
       "      <td>-0.009834</td>\n",
       "      <td>0.101132</td>\n",
       "      <td>-0.124272</td>\n",
       "      <td>0.116321</td>\n",
       "      <td>-0.053066</td>\n",
       "      <td>-0.072625</td>\n",
       "      <td>0.010508</td>\n",
       "      <td>0.011203</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.282446</td>\n",
       "      <td>-0.128331</td>\n",
       "      <td>1.502726</td>\n",
       "      <td>-0.618458</td>\n",
       "      <td>-0.902702</td>\n",
       "      <td>-0.011136</td>\n",
       "      <td>-0.228185</td>\n",
       "      <td>-0.291438</td>\n",
       "      <td>0.214832</td>\n",
       "      <td>1.020817</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.105812</td>\n",
       "      <td>0.055736</td>\n",
       "      <td>0.166189</td>\n",
       "      <td>-0.151427</td>\n",
       "      <td>0.598960</td>\n",
       "      <td>0.020242</td>\n",
       "      <td>-0.023610</td>\n",
       "      <td>-0.023072</td>\n",
       "      <td>-0.180625</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-0.085061</td>\n",
       "      <td>0.102229</td>\n",
       "      <td>0.840554</td>\n",
       "      <td>-0.730703</td>\n",
       "      <td>-1.359444</td>\n",
       "      <td>-0.167428</td>\n",
       "      <td>-0.583728</td>\n",
       "      <td>-0.231280</td>\n",
       "      <td>0.236922</td>\n",
       "      <td>1.143307</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.152038</td>\n",
       "      <td>-0.007882</td>\n",
       "      <td>0.119603</td>\n",
       "      <td>-0.060771</td>\n",
       "      <td>0.566385</td>\n",
       "      <td>0.077041</td>\n",
       "      <td>-0.069192</td>\n",
       "      <td>-0.018181</td>\n",
       "      <td>-0.052448</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 56 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        PC1       PC2       PC3       PC4       PC5       PC6       PC7  \\\n",
       "0 -0.121848 -0.109631  0.677826  1.666928  1.618763 -0.550915  0.724746   \n",
       "1 -0.294837 -0.135038  0.612546 -0.003972  0.179251  0.200744  0.551912   \n",
       "2 -0.296454 -0.122306  0.340623 -0.655986 -1.082754  0.059316 -0.495534   \n",
       "3 -0.282446 -0.128331  1.502726 -0.618458 -0.902702 -0.011136 -0.228185   \n",
       "4 -0.085061  0.102229  0.840554 -0.730703 -1.359444 -0.167428 -0.583728   \n",
       "\n",
       "        PC8       PC9      PC10  ...      PC47      PC48      PC49      PC50  \\\n",
       "0 -0.201950 -0.829159 -0.149451  ...  0.065499  0.062655 -0.133934  0.168381   \n",
       "1  0.277117  0.068057 -1.479715  ...  0.201665 -0.068814 -0.228309  0.154042   \n",
       "2 -0.429369  0.711304 -0.653683  ... -0.023068 -0.009834  0.101132 -0.124272   \n",
       "3 -0.291438  0.214832  1.020817  ... -0.105812  0.055736  0.166189 -0.151427   \n",
       "4 -0.231280  0.236922  1.143307  ... -0.152038 -0.007882  0.119603 -0.060771   \n",
       "\n",
       "       PC51      PC52      PC53      PC54      PC55  Malware  \n",
       "0 -0.588060  0.035701  0.005702  0.009438  0.073707        1  \n",
       "1 -0.806254 -0.028854 -0.015292  0.066456  0.122343        1  \n",
       "2  0.116321 -0.053066 -0.072625  0.010508  0.011203        1  \n",
       "3  0.598960  0.020242 -0.023610 -0.023072 -0.180625        1  \n",
       "4  0.566385  0.077041 -0.069192 -0.018181 -0.052448        1  \n",
       "\n",
       "[5 rows x 56 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "features = df.drop(columns=['Malware', 'Name', 'TimeDateStamp', 'Machine'])\n",
    "# We have removed these parameters because they present general information\n",
    "#     and not an essential parameter of the file itself\n",
    "\n",
    "scaler = StandardScaler()\n",
    "features_standardized = scaler.fit_transform(features)\n",
    "\n",
    "pca = PCA(n_components=55)  # 55 - following the research paper's number of dimensions\n",
    "features_pca = pca.fit_transform(features_standardized)\n",
    "\n",
    "# Create a new DataFrame with the selected principal components and the target label\n",
    "columns_pca = [f'PC{i+1}' for i in range(features_pca.shape[1])]\n",
    "df_pca = pd.DataFrame(data=features_pca, columns=columns_pca)\n",
    "df_pca['Malware'] = df['Malware']  # Add the label column \n",
    "print(df_pca.shape)\n",
    "df_pca.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1712aa59",
   "metadata": {},
   "source": [
    "## Models Building:\n",
    "- Gaussian Naïve Bayes,\n",
    "- Decision Tree,\n",
    "- Random Forest, \n",
    "- AdaBoost, \n",
    "- Gradient Boosting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "de55ebeb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train size: (15688, 55)\n",
      "y_train size: (15688,) \n",
      "\n",
      "X_test size: (3923, 55)\n",
      "y_test size: (3923,)\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(df_pca.drop(\"Malware\", axis=1), df[\"Malware\"],\n",
    "                                                    test_size=0.2, random_state=42)\n",
    "print(\"X_train size:\", X_train.shape)\n",
    "print(\"y_train size:\", y_train.shape,\"\\n\")\n",
    "print(\"X_test size:\", X_test.shape)\n",
    "print(\"y_test size:\", y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "196cda61",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 0 1 ... 0 0 0]\n",
      "[1 1 0 ... 1 1 1]\n",
      "Naive Bayes Recall score:0.93955\n",
      "Decision Tree Recall score:0.97985\n",
      "Random Forest Recall score:0.99112\n",
      "AdaBoost Recall score:0.98156\n",
      "Gradient Boosting Recall score:0.98839\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1400x600 with 11 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier, AdaBoostClassifier\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import ConfusionMatrixDisplay\n",
    "from sklearn.metrics import recall_score, accuracy_score, precision_score\n",
    "\n",
    "# Initialize classifiers\n",
    "nb_classifier = GaussianNB()\n",
    "dt_classifier = DecisionTreeClassifier(random_state=42)\n",
    "rf_classifier = RandomForestClassifier(random_state=42)\n",
    "ab_classifier = AdaBoostClassifier(random_state=42)\n",
    "gb_classifier = GradientBoostingClassifier(random_state=42)\n",
    "\n",
    "classifiers = [\n",
    "    ('Naive Bayes', nb_classifier),\n",
    "    ('Decision Tree', dt_classifier),\n",
    "    ('Random Forest', rf_classifier),\n",
    "    ('AdaBoost', ab_classifier),\n",
    "    ('Gradient Boosting', gb_classifier)\n",
    "]\n",
    "\n",
    "# Set up subplots\n",
    "n_rows = (len(classifiers) + 1) // 2\n",
    "n_cols = 2\n",
    "fig, axs = plt.subplots(n_rows, n_cols, figsize=(14, 6))\n",
    "\n",
    "# Flatten the axs array if it's more than one row\n",
    "axs = axs.flatten()\n",
    "\n",
    "# Train and evaluate each classifier\n",
    "for idx, (clf_name, clf) in enumerate(classifiers):\n",
    "    clf.fit(X_train, y_train)\n",
    "    y_pred = clf.predict(X_test)\n",
    "    if clf_name == 'Naive Bayes':\n",
    "        print(y_pred)\n",
    "        y_pred = (y_pred+1)%2\n",
    "        print(y_pred)\n",
    "    \n",
    "    # Get confusion matrix\n",
    "    conf_mat = confusion_matrix(y_test, y_pred)\n",
    "    \n",
    "    # Plot confusion matrix in the corresponding subplot\n",
    "    ax = axs[idx]\n",
    "    ConfusionMatrixDisplay(conf_mat, display_labels=['Benign', 'Malware']).plot(ax=ax, cmap=plt.cm.Blues)\n",
    "    ax.set_title(f'Confusion Matrix - {clf_name}')\n",
    "    print(f\"{clf_name} Recall score:{recall_score(y_test, y_pred):.5f}\")\n",
    "\n",
    "# Adjust layout and show the plot\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "717f63da",
   "metadata": {},
   "source": [
    "## Deep learning models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "7aba8d1b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mlp_dense1 Accuracy: 0.9780780015294418\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\ohad1\\anaconda3\\lib\\site-packages\\sklearn\\neural_network\\_multilayer_perceptron.py:614: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (400) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mlp_dense2 Accuracy: 0.9785878154473617\n",
      "Epoch 1/50\n",
      "491/491 [==============================] - 2s 3ms/step - loss: 0.1548 - accuracy: 0.9530\n",
      "Epoch 2/50\n",
      "491/491 [==============================] - 2s 3ms/step - loss: 0.1090 - accuracy: 0.9666\n",
      "Epoch 3/50\n",
      "491/491 [==============================] - 2s 3ms/step - loss: 0.0991 - accuracy: 0.9689\n",
      "Epoch 4/50\n",
      "491/491 [==============================] - 2s 3ms/step - loss: 0.0949 - accuracy: 0.9704\n",
      "Epoch 5/50\n",
      "491/491 [==============================] - 2s 4ms/step - loss: 0.0861 - accuracy: 0.9735\n",
      "Epoch 6/50\n",
      "491/491 [==============================] - 4s 7ms/step - loss: 0.0874 - accuracy: 0.9728\n",
      "Epoch 7/50\n",
      "491/491 [==============================] - 3s 6ms/step - loss: 0.0845 - accuracy: 0.9734\n",
      "Epoch 8/50\n",
      "491/491 [==============================] - 4s 8ms/step - loss: 0.0782 - accuracy: 0.9746\n",
      "Epoch 9/50\n",
      "491/491 [==============================] - 4s 8ms/step - loss: 0.0765 - accuracy: 0.9753\n",
      "Epoch 10/50\n",
      "491/491 [==============================] - 3s 6ms/step - loss: 0.0757 - accuracy: 0.9754\n",
      "Epoch 11/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0734 - accuracy: 0.9767\n",
      "Epoch 12/50\n",
      "491/491 [==============================] - 4s 7ms/step - loss: 0.0705 - accuracy: 0.9775\n",
      "Epoch 13/50\n",
      "491/491 [==============================] - 4s 7ms/step - loss: 0.0672 - accuracy: 0.9783\n",
      "Epoch 14/50\n",
      "491/491 [==============================] - 3s 6ms/step - loss: 0.0676 - accuracy: 0.9779\n",
      "Epoch 15/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0664 - accuracy: 0.9778\n",
      "Epoch 16/50\n",
      "491/491 [==============================] - 3s 7ms/step - loss: 0.0628 - accuracy: 0.9799\n",
      "Epoch 17/50\n",
      "491/491 [==============================] - 5s 11ms/step - loss: 0.0635 - accuracy: 0.9787\n",
      "Epoch 18/50\n",
      "491/491 [==============================] - 4s 9ms/step - loss: 0.0628 - accuracy: 0.9803\n",
      "Epoch 19/50\n",
      "491/491 [==============================] - 3s 7ms/step - loss: 0.0590 - accuracy: 0.9806\n",
      "Epoch 20/50\n",
      "491/491 [==============================] - 3s 7ms/step - loss: 0.0607 - accuracy: 0.9807\n",
      "Epoch 21/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0581 - accuracy: 0.9809\n",
      "Epoch 22/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0590 - accuracy: 0.9809\n",
      "Epoch 23/50\n",
      "491/491 [==============================] - 3s 6ms/step - loss: 0.0559 - accuracy: 0.9807\n",
      "Epoch 24/50\n",
      "491/491 [==============================] - 3s 6ms/step - loss: 0.0521 - accuracy: 0.9823\n",
      "Epoch 25/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0551 - accuracy: 0.9818\n",
      "Epoch 26/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0521 - accuracy: 0.9839\n",
      "Epoch 27/50\n",
      "491/491 [==============================] - 2s 5ms/step - loss: 0.0523 - accuracy: 0.9830\n",
      "Epoch 28/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0507 - accuracy: 0.9845\n",
      "Epoch 29/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0482 - accuracy: 0.9844\n",
      "Epoch 30/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0483 - accuracy: 0.9840\n",
      "Epoch 31/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0514 - accuracy: 0.9827\n",
      "Epoch 32/50\n",
      "491/491 [==============================] - 2s 5ms/step - loss: 0.0479 - accuracy: 0.9838\n",
      "Epoch 33/50\n",
      "491/491 [==============================] - 2s 5ms/step - loss: 0.0453 - accuracy: 0.9844\n",
      "Epoch 34/50\n",
      "491/491 [==============================] - 2s 5ms/step - loss: 0.0465 - accuracy: 0.9854\n",
      "Epoch 35/50\n",
      "491/491 [==============================] - 2s 5ms/step - loss: 0.0533 - accuracy: 0.9844\n",
      "Epoch 36/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0476 - accuracy: 0.9843\n",
      "Epoch 37/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0441 - accuracy: 0.9853\n",
      "Epoch 38/50\n",
      "491/491 [==============================] - 3s 6ms/step - loss: 0.0461 - accuracy: 0.9848\n",
      "Epoch 39/50\n",
      "491/491 [==============================] - 3s 6ms/step - loss: 0.0416 - accuracy: 0.9860\n",
      "Epoch 40/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0448 - accuracy: 0.9854\n",
      "Epoch 41/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0410 - accuracy: 0.9866\n",
      "Epoch 42/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0430 - accuracy: 0.9869\n",
      "Epoch 43/50\n",
      "491/491 [==============================] - 2s 5ms/step - loss: 0.0419 - accuracy: 0.9862\n",
      "Epoch 44/50\n",
      "491/491 [==============================] - 2s 5ms/step - loss: 0.0404 - accuracy: 0.9864\n",
      "Epoch 45/50\n",
      "491/491 [==============================] - 2s 5ms/step - loss: 0.0403 - accuracy: 0.9867\n",
      "Epoch 46/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0407 - accuracy: 0.9865\n",
      "Epoch 47/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0381 - accuracy: 0.9876\n",
      "Epoch 48/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0371 - accuracy: 0.9871\n",
      "Epoch 49/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0424 - accuracy: 0.9862\n",
      "Epoch 50/50\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.0416 - accuracy: 0.9866\n",
      "model_1d_cnn Accuracy: 0.9790976047515869\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score, StratifiedKFold\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "import numpy as np\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Conv1D, MaxPooling1D, Flatten, Dense\n",
    "from tensorflow.keras.losses import SparseCategoricalCrossentropy\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "\n",
    "# Define StratifiedKFold\n",
    "kf = StratifiedKFold(n_splits=10, random_state=42, shuffle=True)\n",
    "\n",
    "# Define MLP models\n",
    "mlp_dense1 = MLPClassifier(hidden_layer_sizes=(100,), activation='tanh', solver='adam',\n",
    "                           learning_rate_init=0.01, alpha=0.0001, batch_size=64,\n",
    "                           random_state=42, validation_fraction=0.2, max_iter=300,\n",
    "                           n_iter_no_change=10)\n",
    "\n",
    "mlp_dense2 = MLPClassifier(hidden_layer_sizes=(200, 100), activation='logistic', solver='adam',\n",
    "                           learning_rate_init=0.001, alpha=0.0001, batch_size=128,\n",
    "                           random_state=42, validation_fraction=0.2, max_iter=400,\n",
    "                           n_iter_no_change=10)\n",
    "\n",
    "# Train and evaluate MLP models\n",
    "for model, name in zip([mlp_dense1, mlp_dense2], ['mlp_dense1', 'mlp_dense2']):\n",
    "    scores = cross_val_score(model, X_train, y_train, cv=kf)\n",
    "    model.fit(X_train, y_train)\n",
    "    accuracy = model.score(X_test, y_test)\n",
    "    print(f\"{name} Accuracy:\", accuracy)\n",
    "\n",
    "# Define 1D CNN model\n",
    "model_1d_cnn = Sequential([\n",
    "    Conv1D(filters=60, kernel_size=3, activation='relu', input_shape=(X_train.shape[1], 1)),\n",
    "    MaxPooling1D(pool_size=2),\n",
    "    Flatten(),\n",
    "    Dense(60, activation='relu'),\n",
    "    Dense(2, activation='softmax')\n",
    "])\n",
    "\n",
    "# Compile and train 1D CNN model\n",
    "model_1d_cnn.compile(optimizer=Adam(), loss=SparseCategoricalCrossentropy(), metrics=['accuracy'])\n",
    "model_1d_cnn.fit(X_train.values.reshape(X_train.shape[0], X_train.shape[1], 1), y_train, epochs=50, verbose=1)\n",
    "\n",
    "# Evaluate 1D CNN model\n",
    "test_loss, test_accuracy = model_1d_cnn.evaluate(X_test.values.reshape(X_test.shape[0], X_test.shape[1], 1), y_test, verbose=0)\n",
    "print(\"model_1d_cnn Accuracy:\", test_accuracy)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c06e7831",
   "metadata": {},
   "source": [
    "### improve the 1D CNN model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "26170005",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "491/491 [==============================] - 3s 5ms/step - loss: 0.1416 - accuracy: 0.9526 - val_loss: 0.1007 - val_accuracy: 0.9732\n",
      "Epoch 2/10\n",
      "491/491 [==============================] - 2s 5ms/step - loss: 0.0935 - accuracy: 0.9720 - val_loss: 0.0916 - val_accuracy: 0.9750\n",
      "Epoch 3/10\n",
      "491/491 [==============================] - 4s 8ms/step - loss: 0.0835 - accuracy: 0.9736 - val_loss: 0.0818 - val_accuracy: 0.9765\n",
      "Epoch 4/10\n",
      "491/491 [==============================] - 4s 7ms/step - loss: 0.0793 - accuracy: 0.9749 - val_loss: 0.0861 - val_accuracy: 0.9740\n",
      "Epoch 5/10\n",
      "491/491 [==============================] - 3s 7ms/step - loss: 0.0703 - accuracy: 0.9766 - val_loss: 0.0866 - val_accuracy: 0.9750\n",
      "Epoch 6/10\n",
      "491/491 [==============================] - 3s 7ms/step - loss: 0.0682 - accuracy: 0.9776 - val_loss: 0.0816 - val_accuracy: 0.9758\n",
      "Epoch 7/10\n",
      "491/491 [==============================] - 3s 7ms/step - loss: 0.0638 - accuracy: 0.9790 - val_loss: 0.0736 - val_accuracy: 0.9786\n",
      "Epoch 8/10\n",
      "491/491 [==============================] - 3s 7ms/step - loss: 0.0596 - accuracy: 0.9806 - val_loss: 0.0756 - val_accuracy: 0.9773\n",
      "Epoch 9/10\n",
      "491/491 [==============================] - 4s 8ms/step - loss: 0.0557 - accuracy: 0.9816 - val_loss: 0.0821 - val_accuracy: 0.9776\n",
      "Epoch 10/10\n",
      "491/491 [==============================] - 4s 7ms/step - loss: 0.0564 - accuracy: 0.9816 - val_loss: 0.0764 - val_accuracy: 0.9799\n",
      "123/123 [==============================] - 0s 3ms/step - loss: 0.0764 - accuracy: 0.9799\n",
      "Accuracy: 0.9798623323440552\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "from tensorflow.keras.models import Sequential\n",
    "from tensorflow.keras.layers import Conv1D, MaxPooling1D, Flatten, Dense\n",
    "\n",
    "# Define the model\n",
    "model = Sequential([\n",
    "    Conv1D(filters=32, kernel_size=3, activation='tanh', input_shape=(X_train.shape[1], 1)),\n",
    "    MaxPooling1D(pool_size=2),\n",
    "    Conv1D(filters=64, kernel_size=3, activation='tanh'),\n",
    "    MaxPooling1D(pool_size=2),\n",
    "    Flatten(),\n",
    "    Dense(128, activation='relu'),\n",
    "    Dense(1, activation='sigmoid')  # Changed output layer to a single neuron with sigmoid activation\n",
    "])\n",
    "\n",
    "# Compile the model\n",
    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])\n",
    "\n",
    "# Reshape the input data\n",
    "X_train_reshaped = X_train.values.reshape(X_train.shape[0], X_train.shape[1], 1)\n",
    "X_test_reshaped = X_test.values.reshape(X_test.shape[0], X_test.shape[1], 1)\n",
    "\n",
    "# Train the model\n",
    "model.fit(X_train_reshaped, y_train, epochs=10, batch_size=32, validation_data=(X_test_reshaped, y_test))\n",
    "\n",
    "# Evaluate the model\n",
    "loss, accuracy = model.evaluate(X_test_reshaped, y_test)\n",
    "print(\"Accuracy:\", accuracy)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "6b1d273d",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-26-ecfa0c66eda0>:20: DeprecationWarning: KerasClassifier is deprecated, use Sci-Keras (https://github.com/adriangb/scikeras) instead. See https://www.adriangb.com/scikeras/stable/migration.html for help migrating.\n",
      "  model = KerasClassifier(build_fn=create_model, epochs=10, batch_size=32, verbose=0)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 3 folds for each of 12 candidates, totalling 36 fits\n",
      "Parameters: {'activation': 'relu', 'loss': 'binary_crossentropy', 'neurons': 40, 'optimizer': 'adam'}\n",
      "Mean Test Score: 0.9720167915026346\n",
      "----------------------\n",
      "Parameters: {'activation': 'relu', 'loss': 'binary_crossentropy', 'neurons': 40, 'optimizer': 'sgd'}\n",
      "Mean Test Score: 0.9629653890927633\n",
      "----------------------\n",
      "Parameters: {'activation': 'relu', 'loss': 'binary_crossentropy', 'neurons': 64, 'optimizer': 'adam'}\n",
      "Mean Test Score: 0.9694670041402181\n",
      "----------------------\n",
      "Parameters: {'activation': 'relu', 'loss': 'binary_crossentropy', 'neurons': 64, 'optimizer': 'sgd'}\n",
      "Mean Test Score: 0.9623279372851054\n",
      "----------------------\n",
      "Parameters: {'activation': 'relu', 'loss': 'binary_crossentropy', 'neurons': 128, 'optimizer': 'adam'}\n",
      "Mean Test Score: 0.9734830260276794\n",
      "----------------------\n",
      "Parameters: {'activation': 'relu', 'loss': 'binary_crossentropy', 'neurons': 128, 'optimizer': 'sgd'}\n",
      "Mean Test Score: 0.9635391434033712\n",
      "----------------------\n",
      "Parameters: {'activation': 'tanh', 'loss': 'binary_crossentropy', 'neurons': 40, 'optimizer': 'adam'}\n",
      "Mean Test Score: 0.9723991751670837\n",
      "----------------------\n",
      "Parameters: {'activation': 'tanh', 'loss': 'binary_crossentropy', 'neurons': 40, 'optimizer': 'sgd'}\n",
      "Mean Test Score: 0.9636664589246114\n",
      "----------------------\n",
      "Parameters: {'activation': 'tanh', 'loss': 'binary_crossentropy', 'neurons': 64, 'optimizer': 'adam'}\n",
      "Mean Test Score: 0.9727180600166321\n",
      "----------------------\n",
      "Parameters: {'activation': 'tanh', 'loss': 'binary_crossentropy', 'neurons': 64, 'optimizer': 'sgd'}\n",
      "Mean Test Score: 0.9629654089609782\n",
      "----------------------\n",
      "Parameters: {'activation': 'tanh', 'loss': 'binary_crossentropy', 'neurons': 128, 'optimizer': 'adam'}\n",
      "Mean Test Score: 0.9722718397776285\n",
      "----------------------\n",
      "Parameters: {'activation': 'tanh', 'loss': 'binary_crossentropy', 'neurons': 128, 'optimizer': 'sgd'}\n",
      "Mean Test Score: 0.9634754061698914\n",
      "----------------------\n",
      "Best: 0.973483 using {'activation': 'relu', 'loss': 'binary_crossentropy', 'neurons': 128, 'optimizer': 'adam'}\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from tensorflow.keras.wrappers.scikit_learn import KerasClassifier\n",
    "\n",
    "# Function to create model\n",
    "def create_model(activation='relu', neurons=128, loss='binary_crossentropy', optimizer='adam'):\n",
    "    model = Sequential([\n",
    "        Conv1D(filters=32, kernel_size=3, activation=activation, input_shape=(X_train.shape[1], 1)),\n",
    "        MaxPooling1D(pool_size=2),\n",
    "        Conv1D(filters=64, kernel_size=3, activation=activation),\n",
    "        MaxPooling1D(pool_size=2),\n",
    "        Flatten(),\n",
    "        Dense(neurons, activation=activation),\n",
    "        Dense(1, activation='sigmoid')\n",
    "    ])\n",
    "    model.compile(optimizer=optimizer, loss=loss, metrics=['accuracy'])\n",
    "    return model\n",
    "\n",
    "# Create KerasClassifier wrapper for use in scikit-learn GridSearchCV\n",
    "model = KerasClassifier(build_fn=create_model, epochs=10, batch_size=32, verbose=0)\n",
    "\n",
    "# Define the parameters to search\n",
    "param_grid = {\n",
    "    'activation': ['relu', 'tanh'],\n",
    "    'neurons': [40, 64, 128],\n",
    "    'loss': ['binary_crossentropy'],\n",
    "    'optimizer': ['adam', 'sgd']\n",
    "}\n",
    "\n",
    "# Perform GridSearchCV\n",
    "grid_search = GridSearchCV(estimator=model, param_grid=param_grid, cv=3, verbose=1)\n",
    "grid_search_result = grid_search.fit(X_train_reshaped, y_train)\n",
    "\n",
    "# Print results after each combination\n",
    "for mean_score, params in zip(grid_search_result.cv_results_['mean_test_score'], grid_search_result.cv_results_['params']):\n",
    "    print(\"Parameters:\", params)\n",
    "    print(\"Mean Test Score:\", mean_score)\n",
    "    print(\"----------------------\")\n",
    "\n",
    "# Print the best parameters and score\n",
    "print(\"Best: %f using %s\" % (grid_search_result.best_score_, grid_search_result.best_params_))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "29ffd7fa",
   "metadata": {},
   "source": [
    "## Ensamble learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "5648ce53",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "new_features: 15688\n",
      "new_features_test: [[0 0 0 0 0]\n",
      " [0 0 0 0 0]\n",
      " [1 1 1 1 1]\n",
      " ...\n",
      " [0 1 1 1 1]\n",
      " [0 1 1 1 1]\n",
      " [0 1 1 1 1]]\n",
      "Ensemble Accuracy: 0.9719602345144023\n"
     ]
    }
   ],
   "source": [
    "# Machine learning models\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.model_selection import cross_val_predict\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(df_pca.drop(\"Malware\", axis=1), df[\"Malware\"],\n",
    "                                                    test_size=0.2, random_state=42)\n",
    "# Step 1:\n",
    "# Select different base and final-stage classifiers\n",
    "classifiers = [\n",
    "    GaussianNB(),\n",
    "    DecisionTreeClassifier(random_state=42),\n",
    "    RandomForestClassifier(random_state=42),\n",
    "    AdaBoostClassifier(random_state=42),\n",
    "    GradientBoostingClassifier(random_state=42)\n",
    "]\n",
    "\n",
    "final_classifier = RandomForestClassifier(random_state=42)\n",
    "\n",
    "\n",
    "# Step 2: Train the Ensemble\n",
    "new_features = []\n",
    "kf = KFold(n_splits=5, random_state=42, shuffle=True)\n",
    "\n",
    "for clf in classifiers:\n",
    "    predictions = cross_val_predict(clf, X_train, y_train, cv=kf)\n",
    "    new_features.append(predictions)\n",
    "\n",
    "new_features = np.array(new_features).T  # Shape will be (number of samples, number of classifiers)\n",
    "print(\"new_features:\",len(new_features))\n",
    "\n",
    "final_classifier.fit(new_features, y_train)\n",
    "\n",
    "# Step 3: Test on New Data\n",
    "new_features_test = []\n",
    "\n",
    "for clf in classifiers:\n",
    "    predictions_test = cross_val_predict(clf, X_test, y_test, cv=kf)\n",
    "    new_features_test.append(predictions_test)\n",
    "\n",
    "new_features_test = np.array(new_features_test).T\n",
    "print(\"new_features_test:\",new_features_test)\n",
    "\n",
    "ensemble_predictions = final_classifier.predict(new_features_test)\n",
    "ensemble_accuracy = accuracy_score(y_test, ensemble_predictions)\n",
    "print(\"Ensemble Accuracy:\", ensemble_accuracy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "f1fb7c01",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Deep learning methods\n",
    "\n",
    "from tensorflow.keras.layers import Dropout, PReLU\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "def train_models(X_train, y_train):\n",
    "    # Define base classifiers\n",
    "    dense1 = Sequential([\n",
    "        Dense(55),\n",
    "        PReLU(),\n",
    "        Dropout(0.3),\n",
    "        Dense(2, activation='softmax')\n",
    "    ])\n",
    "\n",
    "    dense2 = Sequential([\n",
    "        Dense(64),\n",
    "        PReLU(),\n",
    "        Dropout(0.3),\n",
    "        Dense(32),\n",
    "        PReLU(),\n",
    "        Dropout(0.3),\n",
    "        Dense(2, activation='softmax')\n",
    "    ])\n",
    "\n",
    "    D1_CNN = Sequential([\n",
    "        Conv1D(filters=32, kernel_size=3, activation='relu', input_shape=(55, 1)),\n",
    "        MaxPooling1D(pool_size=2),\n",
    "        Conv1D(filters=64, kernel_size=3, activation='relu'),\n",
    "        MaxPooling1D(pool_size=2),\n",
    "        Flatten(),\n",
    "        Dense(55),\n",
    "        Dropout(0.5),\n",
    "        Dense(2, activation='softmax')\n",
    "    ])\n",
    "\n",
    "    # Train each model\n",
    "    dense1.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n",
    "    dense1.fit(X_train, y_train, epochs=10, batch_size=32, validation_split=0.2)\n",
    "\n",
    "    dense2.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n",
    "    dense2.fit(X_train, y_train, epochs=10, batch_size=32, validation_split=0.2)\n",
    "\n",
    "    X_train_cnn = np.expand_dims(X_train, axis=-1)  # Reshape for 1D CNN\n",
    "    D1_CNN.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n",
    "    D1_CNN.fit(X_train_cnn, y_train, epochs=10, batch_size=32, validation_split=0.2)\n",
    "\n",
    "    # Generate predictions from each model\n",
    "    dense1_predictions = dense1.predict(X_train)\n",
    "    dense2_predictions = dense2.predict(X_train)\n",
    "    D1_CNN_predictions = D1_CNN.predict(X_train_cnn)\n",
    "\n",
    "    # Combine predictions into new features\n",
    "    new_features = np.concatenate((dense1_predictions, dense2_predictions, D1_CNN_predictions), axis=1)\n",
    "    return new_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "009bd4a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier, ExtraTreesClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis, QuadraticDiscriminantAnalysis\n",
    "from sklearn.linear_model import LogisticRegression, PassiveAggressiveClassifier, RidgeClassifier, SGDClassifier\n",
    "from sklearn.ensemble import IsolationForest\n",
    "\n",
    "def test_ensemble(X_test, y_test, new_features):\n",
    "    # Define final-stage classifiers\n",
    "    classifiers = {\n",
    "        'Decision Tree': DecisionTreeClassifier(),\n",
    "        'SVM (Linear Kernel)': SVC(kernel='linear', probability=True),\n",
    "        'SVM (RBF Kernel)': SVC(kernel='rbf', probability=True),\n",
    "        'Random Forest': RandomForestClassifier(),\n",
    "        'AdaBoost': AdaBoostClassifier(),\n",
    "        'Extra Trees': ExtraTreesClassifier(),\n",
    "        'KNN': KNeighborsClassifier(),\n",
    "        'Gaussian NB': GaussianNB(),\n",
    "        'LDA': LinearDiscriminantAnalysis(),\n",
    "        'QDA': QuadraticDiscriminantAnalysis(),\n",
    "        'Logistic Regression': LogisticRegression(),\n",
    "        'Passive Aggressive': PassiveAggressiveClassifier(),\n",
    "        'Ridge Classifier': RidgeClassifier(),\n",
    "        'SGD': SGDClassifier(),\n",
    "        'Isolation Forest': IsolationForest()\n",
    "    }\n",
    "    \n",
    "    sorted_scores = []\n",
    "    # Train each final-stage classifier and evaluate\n",
    "    for name, classifier in classifiers.items():\n",
    "        classifier.fit(new_features, y_train)\n",
    "        try:\n",
    "            accuracy = classifier.score(new_features, y_train)\n",
    "            sorted_scores.append((name,accuracy))\n",
    "            sorted_scores.sort(key=lambda x: x[1], reverse=True)  # Sort by accuracy in descending order\n",
    "#             print(f'{name} Accuracy: {accuracy:.5f}')\n",
    "        except:\n",
    "            continue\n",
    "    return sorted_scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "4cb649f3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "393/393 [==============================] - 2s 3ms/step - loss: 0.2466 - accuracy: 0.9191 - val_loss: 0.1399 - val_accuracy: 0.9640\n",
      "Epoch 2/10\n",
      "393/393 [==============================] - 1s 3ms/step - loss: 0.1370 - accuracy: 0.9628 - val_loss: 0.1190 - val_accuracy: 0.9662\n",
      "Epoch 3/10\n",
      "393/393 [==============================] - 1s 3ms/step - loss: 0.1264 - accuracy: 0.9622 - val_loss: 0.1074 - val_accuracy: 0.9678\n",
      "Epoch 4/10\n",
      "393/393 [==============================] - 1s 3ms/step - loss: 0.1216 - accuracy: 0.9633 - val_loss: 0.1053 - val_accuracy: 0.9685\n",
      "Epoch 5/10\n",
      "393/393 [==============================] - 1s 3ms/step - loss: 0.1115 - accuracy: 0.9667 - val_loss: 0.1024 - val_accuracy: 0.9678\n",
      "Epoch 6/10\n",
      "393/393 [==============================] - 1s 3ms/step - loss: 0.1084 - accuracy: 0.9655 - val_loss: 0.0996 - val_accuracy: 0.9707\n",
      "Epoch 7/10\n",
      "393/393 [==============================] - 1s 4ms/step - loss: 0.1041 - accuracy: 0.9687 - val_loss: 0.0985 - val_accuracy: 0.9691\n",
      "Epoch 8/10\n",
      "393/393 [==============================] - 2s 4ms/step - loss: 0.1025 - accuracy: 0.9684 - val_loss: 0.0979 - val_accuracy: 0.9723\n",
      "Epoch 9/10\n",
      "393/393 [==============================] - 2s 4ms/step - loss: 0.0996 - accuracy: 0.9700 - val_loss: 0.0974 - val_accuracy: 0.9713\n",
      "Epoch 10/10\n",
      "393/393 [==============================] - 2s 5ms/step - loss: 0.0957 - accuracy: 0.9696 - val_loss: 0.0939 - val_accuracy: 0.9729\n",
      "Epoch 1/10\n",
      "393/393 [==============================] - 3s 5ms/step - loss: 0.2364 - accuracy: 0.9219 - val_loss: 0.1251 - val_accuracy: 0.9665\n",
      "Epoch 2/10\n",
      "393/393 [==============================] - 2s 5ms/step - loss: 0.1458 - accuracy: 0.9593 - val_loss: 0.1062 - val_accuracy: 0.9691\n",
      "Epoch 3/10\n",
      "393/393 [==============================] - 2s 5ms/step - loss: 0.1239 - accuracy: 0.9633 - val_loss: 0.1058 - val_accuracy: 0.9685\n",
      "Epoch 4/10\n",
      "393/393 [==============================] - 1s 4ms/step - loss: 0.1154 - accuracy: 0.9655 - val_loss: 0.1047 - val_accuracy: 0.9704\n",
      "Epoch 5/10\n",
      "393/393 [==============================] - 3s 7ms/step - loss: 0.1112 - accuracy: 0.9661 - val_loss: 0.1006 - val_accuracy: 0.9710\n",
      "Epoch 6/10\n",
      "393/393 [==============================] - 3s 7ms/step - loss: 0.1039 - accuracy: 0.9690 - val_loss: 0.1011 - val_accuracy: 0.9742\n",
      "Epoch 7/10\n",
      "393/393 [==============================] - 3s 7ms/step - loss: 0.0998 - accuracy: 0.9688 - val_loss: 0.1010 - val_accuracy: 0.9742\n",
      "Epoch 8/10\n",
      "393/393 [==============================] - 2s 5ms/step - loss: 0.0967 - accuracy: 0.9719 - val_loss: 0.1004 - val_accuracy: 0.9739\n",
      "Epoch 9/10\n",
      "393/393 [==============================] - 2s 4ms/step - loss: 0.0939 - accuracy: 0.9716 - val_loss: 0.0987 - val_accuracy: 0.9748\n",
      "Epoch 10/10\n",
      "393/393 [==============================] - 3s 7ms/step - loss: 0.0922 - accuracy: 0.9705 - val_loss: 0.0963 - val_accuracy: 0.9758\n",
      "Epoch 1/10\n",
      "393/393 [==============================] - 7s 11ms/step - loss: 0.1975 - accuracy: 0.9382 - val_loss: 0.1307 - val_accuracy: 0.9646\n",
      "Epoch 2/10\n",
      "393/393 [==============================] - 5s 12ms/step - loss: 0.1266 - accuracy: 0.9637 - val_loss: 0.1187 - val_accuracy: 0.9675\n",
      "Epoch 3/10\n",
      "393/393 [==============================] - 5s 14ms/step - loss: 0.1180 - accuracy: 0.9666 - val_loss: 0.1239 - val_accuracy: 0.9694\n",
      "Epoch 4/10\n",
      "393/393 [==============================] - 6s 15ms/step - loss: 0.1168 - accuracy: 0.9672 - val_loss: 0.1137 - val_accuracy: 0.9720\n",
      "Epoch 5/10\n",
      "393/393 [==============================] - 3s 9ms/step - loss: 0.1052 - accuracy: 0.9707 - val_loss: 0.1015 - val_accuracy: 0.9729\n",
      "Epoch 6/10\n",
      "393/393 [==============================] - 3s 8ms/step - loss: 0.1043 - accuracy: 0.9712 - val_loss: 0.1124 - val_accuracy: 0.9688\n",
      "Epoch 7/10\n",
      "393/393 [==============================] - 3s 8ms/step - loss: 0.0982 - accuracy: 0.9727 - val_loss: 0.1100 - val_accuracy: 0.9732\n",
      "Epoch 8/10\n",
      "393/393 [==============================] - 3s 8ms/step - loss: 0.1016 - accuracy: 0.9703 - val_loss: 0.0990 - val_accuracy: 0.9739\n",
      "Epoch 9/10\n",
      "393/393 [==============================] - 3s 8ms/step - loss: 0.0897 - accuracy: 0.9731 - val_loss: 0.1084 - val_accuracy: 0.9700\n",
      "Epoch 10/10\n",
      "393/393 [==============================] - 4s 9ms/step - loss: 0.0887 - accuracy: 0.9738 - val_loss: 0.1016 - val_accuracy: 0.9710\n",
      "491/491 [==============================] - 1s 2ms/step\n",
      "491/491 [==============================] - 1s 3ms/step\n",
      "491/491 [==============================] - 2s 4ms/step\n",
      "(15688, 6)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\ohad1\\anaconda3\\lib\\site-packages\\sklearn\\discriminant_analysis.py:808: UserWarning: Variables are collinear\n",
      "  warnings.warn(\"Variables are collinear\")\n"
     ]
    }
   ],
   "source": [
    "new_features = train_models(X_train, y_train)\n",
    "print(new_features.shape)\n",
    "sorted_scores = test_ensemble(X_test, y_test, new_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "8e22af71",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Scores:\n",
      "#1: Decision Tree: 0.99975\n",
      "#2: Random Forest: 0.99975\n",
      "#3: Extra Trees: 0.99975\n",
      "#4: KNN: 0.98413\n",
      "#5: Logistic Regression: 0.97839\n",
      "#6: AdaBoost: 0.97807\n",
      "#7: Ridge Classifier: 0.97775\n",
      "#8: LDA: 0.97763\n",
      "#9: SVM (RBF Kernel): 0.97750\n",
      "#10: SVM (Linear Kernel): 0.97743\n",
      "#11: SGD: 0.97743\n",
      "#12: Passive Aggressive: 0.97489\n",
      "#13: Gaussian NB: 0.97469\n",
      "#14: QDA: 0.96717\n"
     ]
    }
   ],
   "source": [
    "print(\"Best Scores:\")\n",
    "for i, (name, accuracy) in enumerate(sorted_scores, start=1):\n",
    "    print(f'#{i}: {name}: {accuracy:.5f}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "df4a3467",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters for Decision Tree: {'max_depth': 10, 'max_features': 'log2', 'min_samples_leaf': 4, 'min_samples_split': 2}\n",
      "Best parameters for Random Forest: {'max_depth': None, 'max_features': 'sqrt', 'min_samples_leaf': 1, 'min_samples_split': 2, 'n_estimators': 200}\n",
      "Best parameters for Extra Trees: {'max_depth': None, 'max_features': 'log2', 'min_samples_leaf': 1, 'min_samples_split': 2, 'n_estimators': 100}\n",
      "Best score for Decision Tree: 0.9778174029699187\n",
      "Best score for Random Forest: 0.9834266601082945\n",
      "Best score for Extra Trees: 0.9832991294309392\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.ensemble import RandomForestClassifier, ExtraTreesClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "\n",
    "# Define parameter grids for hyperparameter tuning\n",
    "param_grid_dt = {\n",
    "    'max_depth': [None, 10, 20, 30],\n",
    "    'min_samples_split': [2, 5, 10],\n",
    "    'min_samples_leaf': [1, 2, 4],\n",
    "    'max_features': ['auto', 'sqrt', 'log2', None]\n",
    "}\n",
    "\n",
    "param_grid_rf = {\n",
    "    'n_estimators': [50, 100, 200],\n",
    "    'max_depth': [None, 10, 20, 30],\n",
    "    'min_samples_split': [2, 5, 10],\n",
    "    'min_samples_leaf': [1, 2, 4],\n",
    "    'max_features': ['auto', 'sqrt', 'log2']\n",
    "}\n",
    "\n",
    "param_grid_et = {\n",
    "    'n_estimators': [50, 100, 200],\n",
    "    'max_depth': [None, 10, 20, 30],\n",
    "    'min_samples_split': [2, 5, 10],\n",
    "    'min_samples_leaf': [1, 2, 4],\n",
    "    'max_features': ['auto', 'sqrt', 'log2']\n",
    "}\n",
    "\n",
    "# Perform Grid Search for Decision Tree\n",
    "grid_dt = GridSearchCV(DecisionTreeClassifier(), param_grid_dt, cv=5, scoring='accuracy')\n",
    "grid_dt.fit(new_features, y_train)\n",
    "\n",
    "# Perform Grid Search for Random Forest\n",
    "grid_rf = GridSearchCV(RandomForestClassifier(), param_grid_rf, cv=5, scoring='accuracy')\n",
    "grid_rf.fit(new_features, y_train)\n",
    "\n",
    "# Perform Grid Search for Extra Trees\n",
    "grid_et = GridSearchCV(ExtraTreesClassifier(), param_grid_et, cv=5, scoring='accuracy')\n",
    "grid_et.fit(new_features, y_train)\n",
    "\n",
    "# Print best hyperparameters and scores\n",
    "print(\"Best parameters for Decision Tree:\", grid_dt.best_params_)\n",
    "print(\"Best parameters for Random Forest:\", grid_rf.best_params_)\n",
    "print(\"Best parameters for Extra Trees:\", grid_et.best_params_)\n",
    "print(\"Best score for Decision Tree:\", grid_dt.best_score_)\n",
    "print(\"Best score for Random Forest:\", grid_rf.best_score_)\n",
    "print(\"Best score for Extra Trees:\", grid_et.best_score_)"
   ]
  }
 ],
 "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}