notebook03.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "27044d72",
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "execution": {
     "iopub.execute_input": "2023-06-13T08:56:47.605315Z",
     "iopub.status.busy": "2023-06-13T08:56:47.604883Z",
     "iopub.status.idle": "2023-06-13T08:56:49.184002Z",
     "shell.execute_reply": "2023-06-13T08:56:49.183202Z"
    },
    "papermill": {
     "duration": 1.586046,
     "end_time": "2023-06-13T08:56:49.186129",
     "exception": false,
     "start_time": "2023-06-13T08:56:47.600083",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.metrics import silhouette_score\n",
    "from sklearn.preprocessing import OneHotEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler, LabelEncoder\n",
    "from sklearn.decomposition import PCA\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from yellowbrick.cluster import KElbowVisualizer"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f2db5d6",
   "metadata": {},
   "source": [
    "# Anomaly Detection with Clustering\n",
    "\n",
    "[KDDCup99](https://www.kaggle.com/datasets/galaxyh/kdd-cup-1999-data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "32a87f8b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-13T08:56:49.194307Z",
     "iopub.status.busy": "2023-06-13T08:56:49.193332Z",
     "iopub.status.idle": "2023-06-13T08:56:50.796091Z",
     "shell.execute_reply": "2023-06-13T08:56:50.795222Z"
    },
    "papermill": {
     "duration": 1.608646,
     "end_time": "2023-06-13T08:56:50.798203",
     "exception": false,
     "start_time": "2023-06-13T08:56:49.189557",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "data = pd.read_csv('dataset/KDDCup99.csv.gz')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ae916abd",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-13T08:56:50.811175Z",
     "iopub.status.busy": "2023-06-13T08:56:50.810854Z",
     "iopub.status.idle": "2023-06-13T08:56:50.981004Z",
     "shell.execute_reply": "2023-06-13T08:56:50.980010Z"
    },
    "papermill": {
     "duration": 0.176543,
     "end_time": "2023-06-13T08:56:50.983354",
     "exception": false,
     "start_time": "2023-06-13T08:56:50.806811",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>duration</th>\n",
       "      <th>src_bytes</th>\n",
       "      <th>dst_bytes</th>\n",
       "      <th>land</th>\n",
       "      <th>wrong_fragment</th>\n",
       "      <th>urgent</th>\n",
       "      <th>hot</th>\n",
       "      <th>num_failed_logins</th>\n",
       "      <th>logged_in</th>\n",
       "      <th>lnum_compromised</th>\n",
       "      <th>...</th>\n",
       "      <th>dst_host_count</th>\n",
       "      <th>dst_host_srv_count</th>\n",
       "      <th>dst_host_same_srv_rate</th>\n",
       "      <th>dst_host_diff_srv_rate</th>\n",
       "      <th>dst_host_same_src_port_rate</th>\n",
       "      <th>dst_host_srv_diff_host_rate</th>\n",
       "      <th>dst_host_serror_rate</th>\n",
       "      <th>dst_host_srv_serror_rate</th>\n",
       "      <th>dst_host_rerror_rate</th>\n",
       "      <th>dst_host_srv_rerror_rate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>494020.000000</td>\n",
       "      <td>4.940200e+05</td>\n",
       "      <td>4.940200e+05</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "      <td>494020.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>47.979400</td>\n",
       "      <td>3.025616e+03</td>\n",
       "      <td>8.685308e+02</td>\n",
       "      <td>0.000045</td>\n",
       "      <td>0.006433</td>\n",
       "      <td>0.000014</td>\n",
       "      <td>0.034519</td>\n",
       "      <td>0.000152</td>\n",
       "      <td>0.148245</td>\n",
       "      <td>0.010212</td>\n",
       "      <td>...</td>\n",
       "      <td>232.471248</td>\n",
       "      <td>188.666052</td>\n",
       "      <td>0.753781</td>\n",
       "      <td>0.030906</td>\n",
       "      <td>0.601936</td>\n",
       "      <td>0.006684</td>\n",
       "      <td>0.176754</td>\n",
       "      <td>0.176443</td>\n",
       "      <td>0.058118</td>\n",
       "      <td>0.057412</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>707.747185</td>\n",
       "      <td>9.882191e+05</td>\n",
       "      <td>3.304003e+04</td>\n",
       "      <td>0.006673</td>\n",
       "      <td>0.134805</td>\n",
       "      <td>0.005510</td>\n",
       "      <td>0.782103</td>\n",
       "      <td>0.015520</td>\n",
       "      <td>0.355343</td>\n",
       "      <td>1.798328</td>\n",
       "      <td>...</td>\n",
       "      <td>64.744601</td>\n",
       "      <td>106.040205</td>\n",
       "      <td>0.410780</td>\n",
       "      <td>0.109259</td>\n",
       "      <td>0.481309</td>\n",
       "      <td>0.042133</td>\n",
       "      <td>0.380593</td>\n",
       "      <td>0.380920</td>\n",
       "      <td>0.230590</td>\n",
       "      <td>0.230141</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.500000e+01</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>255.000000</td>\n",
       "      <td>46.000000</td>\n",
       "      <td>0.410000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>5.200000e+02</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>255.000000</td>\n",
       "      <td>255.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.032000e+03</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>255.000000</td>\n",
       "      <td>255.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.040000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>58329.000000</td>\n",
       "      <td>6.933756e+08</td>\n",
       "      <td>5.155468e+06</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>884.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>255.000000</td>\n",
       "      <td>255.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 38 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            duration     src_bytes     dst_bytes           land  \\\n",
       "count  494020.000000  4.940200e+05  4.940200e+05  494020.000000   \n",
       "mean       47.979400  3.025616e+03  8.685308e+02       0.000045   \n",
       "std       707.747185  9.882191e+05  3.304003e+04       0.006673   \n",
       "min         0.000000  0.000000e+00  0.000000e+00       0.000000   \n",
       "25%         0.000000  4.500000e+01  0.000000e+00       0.000000   \n",
       "50%         0.000000  5.200000e+02  0.000000e+00       0.000000   \n",
       "75%         0.000000  1.032000e+03  0.000000e+00       0.000000   \n",
       "max     58329.000000  6.933756e+08  5.155468e+06       1.000000   \n",
       "\n",
       "       wrong_fragment         urgent            hot  num_failed_logins  \\\n",
       "count   494020.000000  494020.000000  494020.000000      494020.000000   \n",
       "mean         0.006433       0.000014       0.034519           0.000152   \n",
       "std          0.134805       0.005510       0.782103           0.015520   \n",
       "min          0.000000       0.000000       0.000000           0.000000   \n",
       "25%          0.000000       0.000000       0.000000           0.000000   \n",
       "50%          0.000000       0.000000       0.000000           0.000000   \n",
       "75%          0.000000       0.000000       0.000000           0.000000   \n",
       "max          3.000000       3.000000      30.000000           5.000000   \n",
       "\n",
       "           logged_in  lnum_compromised  ...  dst_host_count  \\\n",
       "count  494020.000000     494020.000000  ...   494020.000000   \n",
       "mean        0.148245          0.010212  ...      232.471248   \n",
       "std         0.355343          1.798328  ...       64.744601   \n",
       "min         0.000000          0.000000  ...        0.000000   \n",
       "25%         0.000000          0.000000  ...      255.000000   \n",
       "50%         0.000000          0.000000  ...      255.000000   \n",
       "75%         0.000000          0.000000  ...      255.000000   \n",
       "max         1.000000        884.000000  ...      255.000000   \n",
       "\n",
       "       dst_host_srv_count  dst_host_same_srv_rate  dst_host_diff_srv_rate  \\\n",
       "count       494020.000000           494020.000000           494020.000000   \n",
       "mean           188.666052                0.753781                0.030906   \n",
       "std            106.040205                0.410780                0.109259   \n",
       "min              0.000000                0.000000                0.000000   \n",
       "25%             46.000000                0.410000                0.000000   \n",
       "50%            255.000000                1.000000                0.000000   \n",
       "75%            255.000000                1.000000                0.040000   \n",
       "max            255.000000                1.000000                1.000000   \n",
       "\n",
       "       dst_host_same_src_port_rate  dst_host_srv_diff_host_rate  \\\n",
       "count                494020.000000                494020.000000   \n",
       "mean                      0.601936                     0.006684   \n",
       "std                       0.481309                     0.042133   \n",
       "min                       0.000000                     0.000000   \n",
       "25%                       0.000000                     0.000000   \n",
       "50%                       1.000000                     0.000000   \n",
       "75%                       1.000000                     0.000000   \n",
       "max                       1.000000                     1.000000   \n",
       "\n",
       "       dst_host_serror_rate  dst_host_srv_serror_rate  dst_host_rerror_rate  \\\n",
       "count         494020.000000             494020.000000         494020.000000   \n",
       "mean               0.176754                  0.176443              0.058118   \n",
       "std                0.380593                  0.380920              0.230590   \n",
       "min                0.000000                  0.000000              0.000000   \n",
       "25%                0.000000                  0.000000              0.000000   \n",
       "50%                0.000000                  0.000000              0.000000   \n",
       "75%                0.000000                  0.000000              0.000000   \n",
       "max                1.000000                  1.000000              1.000000   \n",
       "\n",
       "       dst_host_srv_rerror_rate  \n",
       "count             494020.000000  \n",
       "mean                   0.057412  \n",
       "std                    0.230141  \n",
       "min                    0.000000  \n",
       "25%                    0.000000  \n",
       "50%                    0.000000  \n",
       "75%                    0.000000  \n",
       "max                    1.000000  \n",
       "\n",
       "[8 rows x 38 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "bfb51909",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-13T08:56:50.991356Z",
     "iopub.status.busy": "2023-06-13T08:56:50.991051Z",
     "iopub.status.idle": "2023-06-13T08:56:52.323089Z",
     "shell.execute_reply": "2023-06-13T08:56:52.322268Z"
    },
    "papermill": {
     "duration": 1.338329,
     "end_time": "2023-06-13T08:56:52.325076",
     "exception": false,
     "start_time": "2023-06-13T08:56:50.986747",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "395216\n"
     ]
    }
   ],
   "source": [
    "Y = data.iloc[:,-1] \n",
    "X = data.iloc[:,:-1]\n",
    "\n",
    "# Categorical features\n",
    "cat_features = ['protocol_type', 'service', 'flag']\n",
    "\n",
    "# One-Hot-Encoding \n",
    "encoder = OneHotEncoder(sparse_output=False)\n",
    "X_encoded = encoder.fit_transform(X[cat_features])\n",
    "\n",
    "new_feature_names = encoder.get_feature_names_out(input_features=cat_features)\n",
    "\n",
    "# Add the encoded array to the data and assign the new column names\n",
    "X_encoded_df = pd.DataFrame(X_encoded, columns=new_feature_names)\n",
    "X = pd.concat([X.drop(columns=cat_features), X_encoded_df], axis=1)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size=0.2, random_state=42)\n",
    "\n",
    "print(len(X_train))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "45bb601d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-13T08:56:52.334361Z",
     "iopub.status.busy": "2023-06-13T08:56:52.333712Z",
     "iopub.status.idle": "2023-06-13T08:57:00.987611Z",
     "shell.execute_reply": "2023-06-13T08:57:00.986697Z"
    },
    "papermill": {
     "duration": 8.660788,
     "end_time": "2023-06-13T08:57:00.989535",
     "exception": false,
     "start_time": "2023-06-13T08:56:52.328747",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x550 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimal number of main components: 64\n"
     ]
    }
   ],
   "source": [
    "# Scaling the data \n",
    "scaler = StandardScaler()\n",
    "X_scaled = scaler.fit_transform(X_train)\n",
    "\n",
    "pca = PCA(random_state=42)\n",
    "pca.fit(X_scaled)\n",
    "\n",
    "explained_variance_ratio = pca.explained_variance_ratio_\n",
    "cumulative_variance = np.cumsum(explained_variance_ratio)\n",
    "\n",
    "plt.plot(range(1, len(cumulative_variance) + 1), cumulative_variance, marker='o')\n",
    "plt.xlabel('Number of main components')\n",
    "plt.ylabel('Cumulative explained variance')\n",
    "plt.title('Cumulative explained variance vs. number of principal components')\n",
    "plt.grid(True)\n",
    "plt.show()\n",
    "\n",
    "threshold = 0.80\n",
    "n_components = np.argmax(cumulative_variance >= threshold) + 1\n",
    "\n",
    "print(f'Optimal number of main components: {n_components}')\n",
    "\n",
    "pca = PCA(n_components=n_components, random_state=42) \n",
    "X_pca = pca.fit_transform(X_scaled)\n",
    "\n",
    "feature_weights = pca.components_\n",
    "scaled_weights = feature_weights.T * np.sqrt(pca.explained_variance_)\n",
    "\n",
    "relevant_features = [] \n",
    "index_selected_features = [] \n",
    "for i, weights in enumerate(scaled_weights):\n",
    "    component_features = []\n",
    "    for j, weight in enumerate(weights):\n",
    "        if abs(weight) > 0.4:  \n",
    "            component_features.append(j)  \n",
    "    relevant_features.append(component_features)\n",
    "\n",
    "for i, component_features in enumerate(relevant_features):\n",
    "    for feature_index in component_features:\n",
    "        index_selected_features.append(feature_index)\n",
    "\n",
    "index_selected_features = list(set(index_selected_features))\n",
    "X_scaled = X_train.iloc[:, index_selected_features]  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "5cf389dc",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-13T08:57:00.999888Z",
     "iopub.status.busy": "2023-06-13T08:57:00.999406Z",
     "iopub.status.idle": "2023-06-13T08:57:42.238724Z",
     "shell.execute_reply": "2023-06-13T08:57:42.237757Z"
    },
    "papermill": {
     "duration": 41.246537,
     "end_time": "2023-06-13T08:57:42.240616",
     "exception": false,
     "start_time": "2023-06-13T08:57:00.994079",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x550 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Axes: title={'center': 'Distortion Score Elbow for KMeans Clustering'}, xlabel='k', ylabel='distortion score'>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Instantiate the clustering model and visualizer\n",
    "model = KMeans(n_init='auto')\n",
    "visualizer = KElbowVisualizer(model, k=(3, 15),  timings=False)\n",
    "\n",
    "visualizer.fit(X_train)  # Fit the data to the visualizer\n",
    "visualizer.show()        # Finalize and render the figure"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "d0152f7c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-13T08:57:42.253080Z",
     "iopub.status.busy": "2023-06-13T08:57:42.252729Z",
     "iopub.status.idle": "2023-06-13T08:57:46.906947Z",
     "shell.execute_reply": "2023-06-13T08:57:46.906006Z"
    },
    "papermill": {
     "duration": 4.662753,
     "end_time": "2023-06-13T08:57:46.909072",
     "exception": false,
     "start_time": "2023-06-13T08:57:42.246319",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "thresholds: [1.59360458e+03 0.00000000e+00 1.26434573e-01 1.08510803e+01\n",
      " 4.35952914e-01 3.42576265e+01]\n",
      "[False False False ... False  True False]\n"
     ]
    }
   ],
   "source": [
    "k = 6\n",
    "kmeans = KMeans(n_clusters=k, n_init='auto')\n",
    "kmeans.fit(X_train)\n",
    "\n",
    "distances = kmeans.transform(X_train)\n",
    "\n",
    "cluster_indices = kmeans.labels_\n",
    "\n",
    "\n",
    "mask = (cluster_indices[:, None] == np.arange(distances.shape[1]))\n",
    "distances[~mask] = 0\n",
    "\n",
    "dictionary = {i: column for i, column in enumerate(zip(*distances))}\n",
    "\n",
    "\n",
    "mean_distances = [sum(values) / len(values) for key, values in dictionary.items()]\n",
    "thresholds = np.array(mean_distances)  # thresholds als NumPy-Array anpassen\n",
    "\n",
    "print(f'thresholds: {thresholds}')\n",
    "\n",
    "outliers = np.any(np.array(distances) > thresholds, axis=1)  \n",
    "print(f'{outliers}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "b22a637f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2023-06-13T08:57:46.921116Z",
     "iopub.status.busy": "2023-06-13T08:57:46.920838Z",
     "iopub.status.idle": "2023-06-13T08:57:46.931415Z",
     "shell.execute_reply": "2023-06-13T08:57:46.930492Z"
    },
    "papermill": {
     "duration": 0.018326,
     "end_time": "2023-06-13T08:57:46.932982",
     "exception": false,
     "start_time": "2023-06-13T08:57:46.914656",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.9212883347664087\n"
     ]
    }
   ],
   "source": [
    "filtered_outlier = y_train[outliers]\n",
    "\n",
    "filtered_labels = filtered_outlier[filtered_outlier != 'normal']\n",
    "\n",
    "count = (len(filtered_outlier) - len(filtered_labels)) / len(filtered_outlier)\n",
    "\n",
    "print(f'Accuracy: {count}')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv",
   "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": "3.10.12"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 69.239289,
   "end_time": "2023-06-13T08:57:47.906004",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2023-06-13T08:56:38.666715",
   "version": "2.4.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}