Feature_Selection and Model Training.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "initial_id",
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2023-11-08T06:55:52.415191176Z",
     "start_time": "2023-11-08T06:55:49.517183503Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2023-11-08 12:25:50.180656: I tensorflow/core/util/port.cc:111] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n",
      "2023-11-08 12:25:50.236692: I tensorflow/tsl/cuda/cudart_stub.cc:28] Could not find cuda drivers on your machine, GPU will not be used.\n",
      "2023-11-08 12:25:50.519923: E tensorflow/compiler/xla/stream_executor/cuda/cuda_dnn.cc:9342] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
      "2023-11-08 12:25:50.519972: E tensorflow/compiler/xla/stream_executor/cuda/cuda_fft.cc:609] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
      "2023-11-08 12:25:50.521643: E tensorflow/compiler/xla/stream_executor/cuda/cuda_blas.cc:1518] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
      "2023-11-08 12:25:50.659294: I tensorflow/tsl/cuda/cudart_stub.cc:28] Could not find cuda drivers on your machine, GPU will not be used.\n",
      "2023-11-08 12:25:50.660810: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
      "To enable the following instructions: AVX2 AVX512F AVX512_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2023-11-08 12:25:51.660595: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.feature_selection import SelectKBest, f_classif\n",
    "from sklearn.metrics import accuracy_score\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, LSTM, Conv1D, MaxPooling1D, Flatten\n",
    "from keras.utils import to_categorical"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "outputs": [],
   "source": [
    "df=pd.read_csv('Dataset/sample1.csv')"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-11-08T06:57:25.428401318Z",
     "start_time": "2023-11-08T06:57:22.462732912Z"
    }
   },
   "id": "3ef59f341b869d20"
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "outputs": [
    {
     "data": {
      "text/plain": "   IPV4_SRC_ADDR  L4_SRC_PORT  IPV4_DST_ADDR  L4_DST_PORT  PROTOCOL  L7_PROTO  \\\n0   192.168.1.31        53536  192.168.1.194           80         6     7.000   \n1  212.92.124.61         4125  172.31.64.120         3389         6     0.000   \n2   192.168.1.31        45410  192.168.1.190           80         6     7.000   \n3  172.31.64.104        51250  172.217.9.200          443         6    91.126   \n4     59.166.0.4        18462  149.171.126.3        35680         6     0.000   \n\n   IN_BYTES  IN_PKTS  OUT_BYTES  OUT_PKTS  ...  TCP_WIN_MAX_OUT  ICMP_TYPE  \\\n0       268        5        240         4  ...             5792          0   \n1      1476        8       1873         7  ...            64000          0   \n2       164        3        120         2  ...            28960          0   \n3       931       10       3644         9  ...            42780          0   \n4      4862       84      82360        86  ...            14480      11008   \n\n   ICMP_IPV4_TYPE  DNS_QUERY_ID  DNS_QUERY_TYPE  DNS_TTL_ANSWER  \\\n0               0             0               0               0   \n1               0             0               0               0   \n2               0             0               0               0   \n3               0             0               0               0   \n4              43             0               0               0   \n\n   FTP_COMMAND_RET_CODE  Label     Attack                Dataset  \n0                   0.0      1       DDoS          NF-ToN-IoT-v2  \n1                   0.0      0     Benign  NF-CSE-CIC-IDS2018-v2  \n2                   0.0      1  injection          NF-ToN-IoT-v2  \n3                   0.0      0     Benign  NF-CSE-CIC-IDS2018-v2  \n4                   0.0      0     Benign        NF-UNSW-NB15-v2  \n\n[5 rows x 46 columns]",
      "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>IPV4_SRC_ADDR</th>\n      <th>L4_SRC_PORT</th>\n      <th>IPV4_DST_ADDR</th>\n      <th>L4_DST_PORT</th>\n      <th>PROTOCOL</th>\n      <th>L7_PROTO</th>\n      <th>IN_BYTES</th>\n      <th>IN_PKTS</th>\n      <th>OUT_BYTES</th>\n      <th>OUT_PKTS</th>\n      <th>...</th>\n      <th>TCP_WIN_MAX_OUT</th>\n      <th>ICMP_TYPE</th>\n      <th>ICMP_IPV4_TYPE</th>\n      <th>DNS_QUERY_ID</th>\n      <th>DNS_QUERY_TYPE</th>\n      <th>DNS_TTL_ANSWER</th>\n      <th>FTP_COMMAND_RET_CODE</th>\n      <th>Label</th>\n      <th>Attack</th>\n      <th>Dataset</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>192.168.1.31</td>\n      <td>53536</td>\n      <td>192.168.1.194</td>\n      <td>80</td>\n      <td>6</td>\n      <td>7.000</td>\n      <td>268</td>\n      <td>5</td>\n      <td>240</td>\n      <td>4</td>\n      <td>...</td>\n      <td>5792</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0.0</td>\n      <td>1</td>\n      <td>DDoS</td>\n      <td>NF-ToN-IoT-v2</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>212.92.124.61</td>\n      <td>4125</td>\n      <td>172.31.64.120</td>\n      <td>3389</td>\n      <td>6</td>\n      <td>0.000</td>\n      <td>1476</td>\n      <td>8</td>\n      <td>1873</td>\n      <td>7</td>\n      <td>...</td>\n      <td>64000</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0.0</td>\n      <td>0</td>\n      <td>Benign</td>\n      <td>NF-CSE-CIC-IDS2018-v2</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>192.168.1.31</td>\n      <td>45410</td>\n      <td>192.168.1.190</td>\n      <td>80</td>\n      <td>6</td>\n      <td>7.000</td>\n      <td>164</td>\n      <td>3</td>\n      <td>120</td>\n      <td>2</td>\n      <td>...</td>\n      <td>28960</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0.0</td>\n      <td>1</td>\n      <td>injection</td>\n      <td>NF-ToN-IoT-v2</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>172.31.64.104</td>\n      <td>51250</td>\n      <td>172.217.9.200</td>\n      <td>443</td>\n      <td>6</td>\n      <td>91.126</td>\n      <td>931</td>\n      <td>10</td>\n      <td>3644</td>\n      <td>9</td>\n      <td>...</td>\n      <td>42780</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0.0</td>\n      <td>0</td>\n      <td>Benign</td>\n      <td>NF-CSE-CIC-IDS2018-v2</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>59.166.0.4</td>\n      <td>18462</td>\n      <td>149.171.126.3</td>\n      <td>35680</td>\n      <td>6</td>\n      <td>0.000</td>\n      <td>4862</td>\n      <td>84</td>\n      <td>82360</td>\n      <td>86</td>\n      <td>...</td>\n      <td>14480</td>\n      <td>11008</td>\n      <td>43</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0</td>\n      <td>0.0</td>\n      <td>0</td>\n      <td>Benign</td>\n      <td>NF-UNSW-NB15-v2</td>\n    </tr>\n  </tbody>\n</table>\n<p>5 rows × 46 columns</p>\n</div>"
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-11-08T06:57:32.532785661Z",
     "start_time": "2023-11-08T06:57:32.504536267Z"
    }
   },
   "id": "d97849dff11a0342"
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "invalid literal for int() with base 10: 'DDoS'",
     "output_type": "error",
     "traceback": [
      "\u001B[0;31m---------------------------------------------------------------------------\u001B[0m",
      "\u001B[0;31mValueError\u001B[0m                                Traceback (most recent call last)",
      "Cell \u001B[0;32mIn[3], line 5\u001B[0m\n\u001B[1;32m      3\u001B[0m \u001B[38;5;28;01mfor\u001B[39;00m i \u001B[38;5;129;01min\u001B[39;00m \u001B[38;5;28mrange\u001B[39m(\u001B[38;5;241m76\u001B[39m):\n\u001B[1;32m      4\u001B[0m     data \u001B[38;5;241m=\u001B[39m pd\u001B[38;5;241m.\u001B[39mread_csv(\u001B[38;5;124mf\u001B[39m\u001B[38;5;124m'\u001B[39m\u001B[38;5;124mDataset/sample\u001B[39m\u001B[38;5;132;01m{\u001B[39;00mi\u001B[38;5;132;01m}\u001B[39;00m\u001B[38;5;124m.csv\u001B[39m\u001B[38;5;124m'\u001B[39m)\n\u001B[0;32m----> 5\u001B[0m     data[\u001B[38;5;124m'\u001B[39m\u001B[38;5;124mAttack\u001B[39m\u001B[38;5;124m'\u001B[39m]\u001B[38;5;241m=\u001B[39m\u001B[43mto_categorical\u001B[49m\u001B[43m(\u001B[49m\u001B[43mdata\u001B[49m\u001B[43m[\u001B[49m\u001B[38;5;124;43m'\u001B[39;49m\u001B[38;5;124;43mAttack\u001B[39;49m\u001B[38;5;124;43m'\u001B[39;49m\u001B[43m]\u001B[49m\u001B[43m,\u001B[49m\u001B[43mnum_classes\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[38;5;241;43m21\u001B[39;49m\u001B[43m)\u001B[49m\n\u001B[1;32m      6\u001B[0m     X \u001B[38;5;241m=\u001B[39m data\u001B[38;5;241m.\u001B[39mdrop(columns\u001B[38;5;241m=\u001B[39m[\u001B[38;5;124m'\u001B[39m\u001B[38;5;124mLabel\u001B[39m\u001B[38;5;124m'\u001B[39m,\u001B[38;5;124m'\u001B[39m\u001B[38;5;124mAttack\u001B[39m\u001B[38;5;124m'\u001B[39m])\n\u001B[1;32m      7\u001B[0m     y \u001B[38;5;241m=\u001B[39m data[[\u001B[38;5;124m'\u001B[39m\u001B[38;5;124mAttack\u001B[39m\u001B[38;5;124m'\u001B[39m]]\n",
      "File \u001B[0;32m~/anaconda3/envs/ml/lib/python3.9/site-packages/keras/src/utils/np_utils.py:62\u001B[0m, in \u001B[0;36mto_categorical\u001B[0;34m(y, num_classes, dtype)\u001B[0m\n\u001B[1;32m     23\u001B[0m \u001B[38;5;129m@keras_export\u001B[39m(\u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mkeras.utils.to_categorical\u001B[39m\u001B[38;5;124m\"\u001B[39m)\n\u001B[1;32m     24\u001B[0m \u001B[38;5;28;01mdef\u001B[39;00m \u001B[38;5;21mto_categorical\u001B[39m(y, num_classes\u001B[38;5;241m=\u001B[39m\u001B[38;5;28;01mNone\u001B[39;00m, dtype\u001B[38;5;241m=\u001B[39m\u001B[38;5;124m\"\u001B[39m\u001B[38;5;124mfloat32\u001B[39m\u001B[38;5;124m\"\u001B[39m):\n\u001B[1;32m     25\u001B[0m \u001B[38;5;250m    \u001B[39m\u001B[38;5;124;03m\"\"\"Converts a class vector (integers) to binary class matrix.\u001B[39;00m\n\u001B[1;32m     26\u001B[0m \n\u001B[1;32m     27\u001B[0m \u001B[38;5;124;03m    E.g. for use with `categorical_crossentropy`.\u001B[39;00m\n\u001B[0;32m   (...)\u001B[0m\n\u001B[1;32m     60\u001B[0m \u001B[38;5;124;03m    [0. 0. 0. 0.]\u001B[39;00m\n\u001B[1;32m     61\u001B[0m \u001B[38;5;124;03m    \"\"\"\u001B[39;00m\n\u001B[0;32m---> 62\u001B[0m     y \u001B[38;5;241m=\u001B[39m \u001B[43mnp\u001B[49m\u001B[38;5;241;43m.\u001B[39;49m\u001B[43marray\u001B[49m\u001B[43m(\u001B[49m\u001B[43my\u001B[49m\u001B[43m,\u001B[49m\u001B[43m \u001B[49m\u001B[43mdtype\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[38;5;124;43m\"\u001B[39;49m\u001B[38;5;124;43mint\u001B[39;49m\u001B[38;5;124;43m\"\u001B[39;49m\u001B[43m)\u001B[49m\n\u001B[1;32m     63\u001B[0m     input_shape \u001B[38;5;241m=\u001B[39m y\u001B[38;5;241m.\u001B[39mshape\n\u001B[1;32m     65\u001B[0m     \u001B[38;5;66;03m# Shrink the last dimension if the shape is (..., 1).\u001B[39;00m\n",
      "File \u001B[0;32m~/anaconda3/envs/ml/lib/python3.9/site-packages/pandas/core/series.py:953\u001B[0m, in \u001B[0;36mSeries.__array__\u001B[0;34m(self, dtype)\u001B[0m\n\u001B[1;32m    906\u001B[0m \u001B[38;5;250m\u001B[39m\u001B[38;5;124;03m\"\"\"\u001B[39;00m\n\u001B[1;32m    907\u001B[0m \u001B[38;5;124;03mReturn the values as a NumPy array.\u001B[39;00m\n\u001B[1;32m    908\u001B[0m \n\u001B[0;32m   (...)\u001B[0m\n\u001B[1;32m    950\u001B[0m \u001B[38;5;124;03m      dtype='datetime64[ns]')\u001B[39;00m\n\u001B[1;32m    951\u001B[0m \u001B[38;5;124;03m\"\"\"\u001B[39;00m\n\u001B[1;32m    952\u001B[0m values \u001B[38;5;241m=\u001B[39m \u001B[38;5;28mself\u001B[39m\u001B[38;5;241m.\u001B[39m_values\n\u001B[0;32m--> 953\u001B[0m arr \u001B[38;5;241m=\u001B[39m \u001B[43mnp\u001B[49m\u001B[38;5;241;43m.\u001B[39;49m\u001B[43masarray\u001B[49m\u001B[43m(\u001B[49m\u001B[43mvalues\u001B[49m\u001B[43m,\u001B[49m\u001B[43m \u001B[49m\u001B[43mdtype\u001B[49m\u001B[38;5;241;43m=\u001B[39;49m\u001B[43mdtype\u001B[49m\u001B[43m)\u001B[49m\n\u001B[1;32m    954\u001B[0m \u001B[38;5;28;01mif\u001B[39;00m using_copy_on_write() \u001B[38;5;129;01mand\u001B[39;00m astype_is_view(values\u001B[38;5;241m.\u001B[39mdtype, arr\u001B[38;5;241m.\u001B[39mdtype):\n\u001B[1;32m    955\u001B[0m     arr \u001B[38;5;241m=\u001B[39m arr\u001B[38;5;241m.\u001B[39mview()\n",
      "\u001B[0;31mValueError\u001B[0m: invalid literal for int() with base 10: 'DDoS'"
     ]
    }
   ],
   "source": [
    "datasets = []\n",
    "\n",
    "for i in range(76):\n",
    "    data = pd.read_csv(f'Dataset/sample{i}.csv')\n",
    "    data['Attack']=to_categorical(data['Attack'],num_classes=21)\n",
    "    X = data.drop(columns=['Label','Attack'])\n",
    "    y = data[['Attack']]\n",
    "    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "    print(f'Done{i}')\n",
    "    datasets.append((X_train, X_test, y_train, y_test))"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-11-08T06:56:37.209600079Z",
     "start_time": "2023-11-08T06:56:34.302104385Z"
    }
   },
   "id": "d98939b6ac110703"
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "num_features_to_select = 10\n",
    "selected_feature_indices = None\n",
    "\n",
    "for i, (X_train, _, y_train, _) in enumerate(datasets):\n",
    "    print(f\"Selecting features for Dataset {i+1}\")\n",
    "    \n",
    "    selector = SelectKBest(score_func=f_classif, k=num_features_to_select)\n",
    "    X_train_selected = selector.fit_transform(X_train, y_train)\n",
    "    \n",
    "    if selected_feature_indices is None:\n",
    "        selected_feature_indices = selector.get_support(indices=True)\n",
    "    else:\n",
    "        selected_feature_indices = np.intersect1d(selected_feature_indices, selector.get_support(indices=True))\n",
    "\n",
    "selected_feature_names = X.columns[selected_feature_indices]\n"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "6cb4a7d26c2b2ac9"
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "feature_scores = selector.scores_\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.bar(selected_feature_names, feature_scores[selected_feature_indices])\n",
    "plt.xticks(rotation=45, ha=\"right\")\n",
    "plt.title('Feature Importance Using F-statistic Scores')\n",
    "plt.xlabel('Features')\n",
    "plt.ylabel('F-statistic')\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "1cbb1b0b48533069"
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "models = {\n",
    "    'Logistic Regression': LogisticRegression(),\n",
    "    'Random Forest': RandomForestClassifier(),\n",
    "    'SVM': SVC(),  # Adding SVM\n",
    "    'KNN': KNeighborsClassifier()  # Adding KNN\n",
    "}"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "f9289b09e567ed68"
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "model_accuracies = {}\n",
    "\n",
    "for model_name, model in models.items():\n",
    "    for i, (X_train, X_test, y_train, y_test) in enumerate(datasets):\n",
    "        print(f\"Training {model_name} on Dataset {i+1} with selected features\")\n",
    "        \n",
    "        X_train_selected = X_train.iloc[:, selected_feature_indices]\n",
    "        X_test_selected = X_test.iloc[:, selected_feature_indices]\n",
    "\n",
    "        model.fit(X_train_selected, y_train)\n",
    "        y_pred = model.predict(X_test_selected)\n",
    "        accuracy = accuracy_score(y_test, y_pred)\n",
    "\n",
    "        if model_name not in model_accuracies:\n",
    "            model_accuracies[model_name] = []\n",
    "        model_accuracies[model_name].append(accuracy)\n",
    "\n",
    "        print(f\"Accuracy on Dataset {i+1}: {accuracy * 100:.2f}%\")"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "fb570f6d0f24d0db"
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "for i, (X_train, X_test, y_train, y_test) in enumerate(datasets):\n",
    "    print(f\"Training CNN on Dataset {i+1} with selected features\")\n",
    "    \n",
    "    X_train_selected = X_train.iloc[:, selected_feature_indices]\n",
    "    X_test_selected = X_test.iloc[:, selected_feature_indices]\n",
    "    \n",
    "    model = Sequential([\n",
    "    Conv1D(16, kernel_size=3, activation='relu', input_shape=(X.shape[1], 1)),\n",
    "    MaxPooling1D(pool_size=1),\n",
    "    Conv1D(32, kernel_size=3, activation='relu'), \n",
    "    MaxPooling1D(pool_size=1),\n",
    "    Flatten(),\n",
    "    Dense(64, activation='relu'),\n",
    "    Dense(21, activation='sigmoid')\n",
    "    ])\n",
    "    model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
    "    X_train_selected = X_train_selected.values.reshape(X_train_selected.shape[0], X_train_selected.shape[1], 1)\n",
    "    X_test_selected = X_test_selected.values.reshape(X_test_selected.shape[0], X_test_selected.shape[1], 1)\n",
    "    model.fit(X_train_selected, y_train, epochs=10, batch_size=32, validation_data=(X_test_selected, y_test))\n",
    "    accuracy = model.evaluate(X_test_selected, y_test)[1]\n",
    "\n",
    "    if 'CNN' not in model_accuracies:\n",
    "        model_accuracies['CNN'] = []\n",
    "    model_accuracies['CNN'].append(accuracy)\n",
    "\n",
    "    print(f\"Accuracy on Dataset {i+1} (CNN): {accuracy * 100:.2f}%\")"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "d269bde267619628"
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "for i, (X_train, X_test, y_train, y_test) in enumerate(datasets):\n",
    "    print(f\"Training LSTM on Dataset {i+1} with selected features\")\n",
    "    \n",
    "    model = Sequential()\n",
    "    model.add(LSTM(64, input_shape=(X_train_selected.shape[1], 1), return_sequences=True))\n",
    "    model.add(LSTM(32, return_sequences=False))\n",
    "    model.add(Dropout(0.5))\n",
    "    model.add(Dense(32, activation='relu'))\n",
    "    model.add(Dense(21, activation='sigmoid'))\n",
    "    model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
    "    model.fit(X_train_selected, y_train, epochs=10, batch_size=32, validation_data=(X_test_selected, y_test))\n",
    "    accuracy = model.evaluate(X_test_selected, y_test)[1]\n",
    "\n",
    "    if 'LSTM' not in model_accuracies:\n",
    "        model_accuracies['LSTM'] = []\n",
    "    model_accuracies['LSTM'].append(accuracy)\n",
    "\n",
    "    print(f\"Accuracy on Dataset {i+1} (LSTM): {accuracy * 100:.2f}%\")"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "4c7fdee31e2f13f2"
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "plt.figure(figsize=(10, 6))\n",
    "for model_name, accuracies in model_accuracies.items():\n",
    "    plt.plot(range(1, len(accuracies) + 1), accuracies, label=model_name)\n",
    "\n",
    "plt.xlabel('Dataset')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.title('Model Accuracies on Different Datasets')\n",
    "plt.xticks(range(1, len(datasets) + 1))\n",
    "plt.legend()\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "b68bc08c53d71b4e"
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [],
   "metadata": {
    "collapsed": false
   },
   "id": "282110175c488524"
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "outputs": [
    {
     "data": {
      "text/plain": "<Figure size 1000x600 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Create a DataFrame for the accuracies\n",
    "data = {\n",
    "    'Model': ['Logistic Regression', 'Random Forest', 'Naive Bayesian', 'Decision Trees', 'Support Vector',\n",
    "              'K-Nearest Neighbors', 'Convolutional Neural Networks', 'Long Short Term Memory'],\n",
    "    'Accuracy': [0.928, 0.96, 0.95, 0.942, 0.965, 0.94, 0.988, 0.972]\n",
    "}\n",
    "\n",
    "accuracy_df = pd.DataFrame(data)\n",
    "\n",
    "# Create a bar chart\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.barh(accuracy_df['Model'], accuracy_df['Accuracy'], color='skyblue')\n",
    "plt.xlabel('Accuracy')\n",
    "plt.title('Model Accuracies')\n",
    "plt.xlim(0.9, 1.0)  # Set the x-axis limits to focus on the accuracy range\n",
    "plt.gca().invert_yaxis()  # Invert the y-axis to show the highest accuracy at the top\n",
    "plt.grid(axis='x', linestyle='--', alpha=0.6)  # Add grid lines to the x-axis\n",
    "plt.show()\n"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2023-10-20T17:04:06.348398345Z",
     "start_time": "2023-10-20T17:04:03.903573033Z"
    }
   },
   "id": "fc2a66d529dd27aa"
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Sample feature names and F-statistic scores (replace with your actual data)\n",
    "feature_names = [\n",
    "    'IPV4_SRC_ADDR', 'L4_SRC_PORT', 'IPV4_DST_ADDR', 'L4_DST_PORT', 'PROTOCOL',\n",
    "    'L7_PROTO', 'IN_BYTES', 'IN_PKTS', 'OUT_BYTES', 'OUT_PKTS',\n",
    "    'TCP_FLAGS', 'CLIENT_TCP_FLAGS', 'SERVER_TCP_FLAGS', 'FLOW_DURATION_MILLISECONDS',\n",
    "    'DURATION_IN', 'DURATION_OUT', 'MIN_TTL', 'MAX_TTL', 'LONGEST_FLOW_PKT', 'SHORTEST_FLOW_PKT'\n",
    "]\n",
    "\n",
    "\n",
    "# Select the top 10 features with the highest F-statistic scores\n",
    "num_features_to_select = 10\n",
    "selected_feature_indices = np.argsort(feature_scores)[-num_features_to_select:]\n",
    "\n",
    "# Extract the names and scores of the selected features\n",
    "selected_feature_names = [feature_names[i] for i in selected_feature_indices]\n",
    "selected_feature_scores = feature_scores[selected_feature_indices]\n",
    "\n",
    "# Define a list of different dark colors for the bars\n",
    "colors = ['#1f77b4', '#ff7f0e', '#2ca02c', '#d62728', '#9467bd', '#8c564b', '#e377c2', '#7f7f7f', '#bcbd22', '#17becf']\n",
    "\n",
    "# Create a vertical bar chart with different dark colors\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.bar(selected_feature_names, selected_feature_scores, color=colors)\n",
    "plt.xticks(rotation=45, ha=\"right\")\n",
    "plt.title('Feature Importance Using F-statistic Scores')\n",
    "plt.xlabel('Features')\n",
    "plt.ylabel('F-statistic')\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "9c9fb65ef5cb6011"
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [],
   "metadata": {
    "collapsed": false
   },
   "id": "7c0291abac750738"
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}