Module 6 Section 1 Lab 1 - K-Nearest Neighbors.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Module 6 Section 1 Lab 1 - K-Nearest Neighbors\n",
    "*k-Nearest Neighbors* (k-NN) merupakan salah satu algoritma *supervised learning* dimana hasil dari *instance* yang baru diklasifikasikan berdasarkan mayoritas dari kategori $k$-tetangga terdekat. k-NN termasuk metode *lazy learning* hanya menyimpan contoh proses pelatihan dan menunda generalisasi (membangun model) hingga *instance* baru harus diklasifikasikan atau dibuat prediksinya. k-NN dapat digunakan untuk tugas klasifikasi dan regresi. k-NN adalah pengklasifikasi berbasis jarak, artinya secara implisit mengasumsikan bahwa semakin kecil jarak antara 2 titik, semakin mirip mereka.\n",
    "\n",
    "KNN memiliki langkah-langkah dasar sebagai berikut:\n",
    "1. Tentukan nilai $k$. Penentuan nilai ini bergantung dari sebaran data. Jika sebaran data rapat dan memiliki *outliers*, maka pilih nilai $k$ yang kecil. Sedangkan jika sebaran data lebar, maka lebih baik memilih nilai $k$ yang besar.\n",
    "2. Hitung jarak.\n",
    "3. Temukan tetangga terdekat. Misalnya, kita menentukan $k=5$. Kita perlu menemukan 5 tetangga terdekat terhadap titik data baru berdasarkan *Euclidean Distance*. \n",
    "4. Lakukan voting. Hitung berapa jumlah data dari masing-masing kategori yang menjadi tetangga (neighbor) titik data baru.\n",
    "5. Klasifikasikan. Masukkan titik data baru ke kategori dengan jumlah data (yang menjadi tetangga titik data baru) terbanyak.\n",
    "\n",
    "<center><img src='https://github.com./ksnugroho/ai-mastery/blob/main/assets/steps-of-knn.png?raw=true' width=600></img></center>\n",
    "\n",
    "## Menentukan Nilai K yang Optimal\n",
    "- Tidak ada metode statistik untuk menemukan nilai $k$ yang paling optimal.\n",
    "- Inisialisasi nilai $k$ secara acak dan mulai komputasi.\n",
    "- Memilih nilai $k$ yang kecil menyebabkan batas keputusan yang tidak stabil.\n",
    "- Kita bisa membuat grafik antara tingkat kesalahan (error rate) dan $k$ yang menunjukkan nilai dalam rentang yang ditentukan. Kemudian pilih nilai $k$ yang memiliki tingkat kesalahan minimum.\n",
    "\n",
    "## Proses Fitting Model\n",
    "k-NN unik dibandingkan dengan algoritma lain karena hampir tidak melakukan apa pun selama langkah proses 'fit', dan semua pekerjaan selama langkah 'prediksi'. Selama langkah 'fit', k-NN hanya menyimpan semua data pelatihan dan nilai yang sesuai. Tidak ada jarak yang dihitung pada titik ini.\n",
    "\n",
    "## Proses Voting\n",
    "Cara memutuskan voting:\n",
    "- Saat melakukan klasifikasi biner, sering gunakan nilai $k$ ganjil untuk menghindari kebingungan menentukan hasilnya.\n",
    "- Untuk masalah klasifikasi multiclass, kurangi $k$ sebanyak 1 untuk melihat siapa yang menang. Ingat ya, bobot suara berdasarkan jarak tetangga.\n",
    "\n",
    "## Notasi\n",
    "Dalam ruang dua dimensi, koordinat diberikan sebagai titik pada sumbu x dan y adalah:\n",
    "\n",
    "$$d=\\sqrt{(x_1-x_2)^2 + (y_1-y_2)^2}$$\n",
    "\n",
    "<center><img src='https://raw.githubusercontent.com/ksnugroho/ai-mastery/main/assets/euclidean-distance.png'></img></center>\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Pregnancies</th>\n",
       "      <th>Glucose</th>\n",
       "      <th>BloodPressure</th>\n",
       "      <th>SkinThickness</th>\n",
       "      <th>Insulin</th>\n",
       "      <th>BMI</th>\n",
       "      <th>DiabetesPedigreeFunction</th>\n",
       "      <th>Age</th>\n",
       "      <th>Outcome</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6</td>\n",
       "      <td>148</td>\n",
       "      <td>72</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>33.6</td>\n",
       "      <td>0.627</td>\n",
       "      <td>50</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>85</td>\n",
       "      <td>66</td>\n",
       "      <td>29</td>\n",
       "      <td>0</td>\n",
       "      <td>26.6</td>\n",
       "      <td>0.351</td>\n",
       "      <td>31</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8</td>\n",
       "      <td>183</td>\n",
       "      <td>64</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>23.3</td>\n",
       "      <td>0.672</td>\n",
       "      <td>32</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>89</td>\n",
       "      <td>66</td>\n",
       "      <td>23</td>\n",
       "      <td>94</td>\n",
       "      <td>28.1</td>\n",
       "      <td>0.167</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>137</td>\n",
       "      <td>40</td>\n",
       "      <td>35</td>\n",
       "      <td>168</td>\n",
       "      <td>43.1</td>\n",
       "      <td>2.288</td>\n",
       "      <td>33</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Pregnancies  Glucose  BloodPressure  SkinThickness  Insulin   BMI  \\\n",
       "0            6      148             72             35        0  33.6   \n",
       "1            1       85             66             29        0  26.6   \n",
       "2            8      183             64              0        0  23.3   \n",
       "3            1       89             66             23       94  28.1   \n",
       "4            0      137             40             35      168  43.1   \n",
       "\n",
       "   DiabetesPedigreeFunction  Age  Outcome  \n",
       "0                     0.627   50        1  \n",
       "1                     0.351   31        0  \n",
       "2                     0.672   32        1  \n",
       "3                     0.167   21        0  \n",
       "4                     2.288   33        1  "
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "data = pd.read_csv('dataset/diabetes.csv')\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = data.iloc[:,:-1]\n",
    "y = data.iloc[:,8:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((576, 8), (192, 8))"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25)\n",
    "X_train.shape, X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Pregnancies</th>\n",
       "      <th>Glucose</th>\n",
       "      <th>BloodPressure</th>\n",
       "      <th>SkinThickness</th>\n",
       "      <th>Insulin</th>\n",
       "      <th>BMI</th>\n",
       "      <th>DiabetesPedigreeFunction</th>\n",
       "      <th>Age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>350</th>\n",
       "      <td>4</td>\n",
       "      <td>92</td>\n",
       "      <td>80</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>42.2</td>\n",
       "      <td>0.237</td>\n",
       "      <td>29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>470</th>\n",
       "      <td>1</td>\n",
       "      <td>144</td>\n",
       "      <td>82</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>41.3</td>\n",
       "      <td>0.607</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>462</th>\n",
       "      <td>8</td>\n",
       "      <td>74</td>\n",
       "      <td>70</td>\n",
       "      <td>40</td>\n",
       "      <td>49</td>\n",
       "      <td>35.3</td>\n",
       "      <td>0.705</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>475</th>\n",
       "      <td>0</td>\n",
       "      <td>137</td>\n",
       "      <td>84</td>\n",
       "      <td>27</td>\n",
       "      <td>0</td>\n",
       "      <td>27.3</td>\n",
       "      <td>0.231</td>\n",
       "      <td>59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>7</td>\n",
       "      <td>133</td>\n",
       "      <td>84</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>40.2</td>\n",
       "      <td>0.696</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>346</th>\n",
       "      <td>1</td>\n",
       "      <td>139</td>\n",
       "      <td>46</td>\n",
       "      <td>19</td>\n",
       "      <td>83</td>\n",
       "      <td>28.7</td>\n",
       "      <td>0.654</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>374</th>\n",
       "      <td>2</td>\n",
       "      <td>122</td>\n",
       "      <td>52</td>\n",
       "      <td>43</td>\n",
       "      <td>158</td>\n",
       "      <td>36.2</td>\n",
       "      <td>0.816</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>216</th>\n",
       "      <td>5</td>\n",
       "      <td>109</td>\n",
       "      <td>62</td>\n",
       "      <td>41</td>\n",
       "      <td>129</td>\n",
       "      <td>35.8</td>\n",
       "      <td>0.514</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>511</th>\n",
       "      <td>0</td>\n",
       "      <td>139</td>\n",
       "      <td>62</td>\n",
       "      <td>17</td>\n",
       "      <td>210</td>\n",
       "      <td>22.1</td>\n",
       "      <td>0.207</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>248</th>\n",
       "      <td>9</td>\n",
       "      <td>124</td>\n",
       "      <td>70</td>\n",
       "      <td>33</td>\n",
       "      <td>402</td>\n",
       "      <td>35.4</td>\n",
       "      <td>0.282</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>576 rows × 8 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Pregnancies  Glucose  BloodPressure  SkinThickness  Insulin   BMI  \\\n",
       "350            4       92             80              0        0  42.2   \n",
       "470            1      144             82             40        0  41.3   \n",
       "462            8       74             70             40       49  35.3   \n",
       "475            0      137             84             27        0  27.3   \n",
       "41             7      133             84              0        0  40.2   \n",
       "..           ...      ...            ...            ...      ...   ...   \n",
       "346            1      139             46             19       83  28.7   \n",
       "374            2      122             52             43      158  36.2   \n",
       "216            5      109             62             41      129  35.8   \n",
       "511            0      139             62             17      210  22.1   \n",
       "248            9      124             70             33      402  35.4   \n",
       "\n",
       "     DiabetesPedigreeFunction  Age  \n",
       "350                     0.237   29  \n",
       "470                     0.607   28  \n",
       "462                     0.705   39  \n",
       "475                     0.231   59  \n",
       "41                      0.696   37  \n",
       "..                        ...  ...  \n",
       "346                     0.654   22  \n",
       "374                     0.816   28  \n",
       "216                     0.514   25  \n",
       "511                     0.207   21  \n",
       "248                     0.282   34  \n",
       "\n",
       "[576 rows x 8 columns]"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [],
   "source": [
    "'''\n",
    "Standardize features by removing the mean and scaling to unit variance.\n",
    "https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html\n",
    "'''\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.preprocessing import MinMaxScaler\n",
    "\n",
    "scaler = StandardScaler()  \n",
    "scaler.fit(X_train)\n",
    "\n",
    "X_train = scaler.transform(X_train)  \n",
    "X_test = scaler.transform(X_test) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.03413896, -0.88274152,  0.55866165, ...,  1.32083043,\n",
       "        -0.71056705, -0.37840264],\n",
       "       [-0.8734322 ,  0.7019078 ,  0.66931816, ...,  1.20107205,\n",
       "         0.41513393, -0.46175252],\n",
       "       [ 1.24423384, -1.43127397,  0.00537914, ...,  0.40268291,\n",
       "         0.71329257,  0.4550961 ],\n",
       "       ...,\n",
       "       [ 0.33666268, -0.36468309, -0.43724688, ...,  0.46921534,\n",
       "         0.13218747, -0.71180214],\n",
       "       [-1.17595592,  0.54953768, -0.43724688, ..., -1.35377322,\n",
       "        -0.8018401 , -1.04520164],\n",
       "       [ 1.54675756,  0.0924273 ,  0.00537914, ...,  0.41598939,\n",
       "        -0.57365747,  0.03834673]])"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Akurasi pada data uji 0.640625 \n",
      "\n",
      "Classification report\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       0.70      0.75      0.73       122\n",
      "           1       0.51      0.44      0.47        70\n",
      "\n",
      "    accuracy                           0.64       192\n",
      "   macro avg       0.61      0.60      0.60       192\n",
      "weighted avg       0.63      0.64      0.63       192\n",
      "\n",
      "Confusion Matrix\n",
      " [[92 30]\n",
      " [39 31]]\n"
     ]
    }
   ],
   "source": [
    "'''\n",
    "Classifier implementing the k-nearest neighbors vote.\n",
    "https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.KNeighborsClassifier.html\n",
    "'''\n",
    "\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.metrics import classification_report, accuracy_score, confusion_matrix\n",
    "\n",
    "knn = KNeighborsClassifier(n_neighbors=1)\n",
    "knn.fit(X_train, y_train.values.ravel())\n",
    "\n",
    "y_pred = knn.predict(X_test)\n",
    "\n",
    "print('Akurasi pada data uji', accuracy_score(y_test, y_pred), '\\n')\n",
    "print('Classification report\\n', classification_report(y_test, y_pred))\n",
    "print('Confusion Matrix\\n', confusion_matrix(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Mencari nilai K optimal berdasarkan akurasi\n",
    "import matplotlib.pyplot as plt  \n",
    "\n",
    "k_range = list(range(1, 20))\n",
    "k_acc = []\n",
    "\n",
    "for k in k_range:\n",
    "    knn = KNeighborsClassifier(n_neighbors=k)\n",
    "    knn.fit(X_train, y_train.values.ravel())\n",
    "\n",
    "    y_pred= knn.predict(X_test)\n",
    "    score = accuracy_score(y_test, y_pred)\n",
    "\n",
    "    k_acc.append(score)\n",
    "\n",
    "plt.figure(figsize=(12, 6))  \n",
    "plt.plot(k_range, k_acc, color='red', linestyle='dashed', marker='o', markerfacecolor='red', markersize=5)\n",
    "plt.title('Accuracy by K Value')  \n",
    "plt.xlabel('K Value')  \n",
    "plt.ylabel('Accuracy') \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Mencari nilai K optimal berdasarkan error rate (kesalahan klasifikasi)\n",
    "\n",
    "error_rate = []\n",
    "\n",
    "for k in k_range:\n",
    "  knn = KNeighborsClassifier(n_neighbors=k)\n",
    "  knn.fit(X_train, y_train.values.ravel())\n",
    "\n",
    "  y_pred = knn.predict(X_test)\n",
    "  y_pred = y_pred.reshape(len(X_test),1)\n",
    "  error_rate.append(np.mean(y_pred!=y_test))\n",
    "\n",
    "plt.figure(figsize=(12, 6))  \n",
    "plt.plot(k_range, error_rate, color='blue', linestyle='dashed', marker='o', markerfacecolor='blue', markersize=5)\n",
    "plt.title('Error Rate by K Value')  \n",
    "plt.xlabel('K Value')  \n",
    "plt.ylabel('Error Rate') \n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Akurasi pada data uji 0.7239583333333334 \n",
      "\n",
      "Classification report\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       0.73      0.90      0.81       122\n",
      "           1       0.71      0.41      0.52        70\n",
      "\n",
      "    accuracy                           0.72       192\n",
      "   macro avg       0.72      0.66      0.66       192\n",
      "weighted avg       0.72      0.72      0.70       192\n",
      "\n",
      "Confusion Matrix\n",
      " [[110  12]\n",
      " [ 41  29]]\n"
     ]
    }
   ],
   "source": [
    "# Retrain dengan nilai K Baru\n",
    "\n",
    "knn_new = KNeighborsClassifier(n_neighbors=19)\n",
    "knn_new.fit(X_train, y_train.values.ravel())\n",
    "\n",
    "y_pred_new = knn_new.predict(X_test)\n",
    "\n",
    "print('Akurasi pada data uji', accuracy_score(y_test, y_pred_new), '\\n')\n",
    "print('Classification report\\n', classification_report(y_test, y_pred_new))\n",
    "print('Confusion Matrix\\n', confusion_matrix(y_test, y_pred_new))"
   ]
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "b3ba2566441a7c06988d0923437866b63cedc61552a5af99d1f4fb67d367b25f"
  },
  "kernelspec": {
   "display_name": "Python 3.8.8 ('base')",
   "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"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}