notebook.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Inisialisasi Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#contoh ada data\n",
    "\n",
    "from os import sep\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "# import k-means\n",
    "from sklearn.cluster import KMeans\n",
    "\n",
    "D = pd.read_csv('seeds_dataset.csv', delimiter=\";\", header=None)\n",
    "D_new = D.copy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Cek Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(210, 9)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "D.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     object\n",
       "1    float64\n",
       "2    float64\n",
       "3    float64\n",
       "4    float64\n",
       "5    float64\n",
       "6    float64\n",
       "7    float64\n",
       "8      int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "D.dtypes\n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math\n",
    "from scipy.special import softmax\n",
    "\n",
    "for j in range(D.shape[1]):\n",
    "    if(D[j].dtype == 'object'):\n",
    "\n",
    "        feature_in_j = D[j].drop_duplicates()\n",
    "        n = feature_in_j.size\n",
    "\n",
    "        get_probability = D.groupby(j).size().div(len(D))\n",
    "\n",
    "        probability = {}\n",
    "        for i in range(feature_in_j.shape[0]):\n",
    "            probability[i] = get_probability[i]\n",
    "\n",
    "        subset_i_of_j = {}\n",
    "        for i in range(n):\n",
    "            subset_i_of_j[feature_in_j.iloc[i]] = 0\n",
    "        \n",
    "        new_subset_j = {}\n",
    "\n",
    "        for i in range(n):\n",
    "            \n",
    "            sum_k_temp = 0\n",
    "            for k in range(i):\n",
    "\n",
    "                p_k = probability[k]\n",
    "                sum_k_temp += (n - k) * p_k\n",
    "            \n",
    "            mu_i = ((n - i) - sum_k_temp) * math.sqrt((1 - (i * probability[i]**3))/ probability[i]*probability[j]*((i - j)**2))\n",
    "\n",
    "            std_dev_i = probability[i]\n",
    "            \n",
    "            gaussian_distribution = np.random.normal(mu_i, std_dev_i)\n",
    "            subset_i_of_j[feature_in_j.iloc[i]] = gaussian_distribution\n",
    "            \n",
    "        D_new[j].replace(subset_i_of_j, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     0      1      2       3      4      5      6      7  8\n",
      "0    a  15.26  14.84  0.8710  5.763  3.312  2.221  5.220  1\n",
      "1    a  14.88  14.57  0.8811  5.554  3.333  1.018  4.956  1\n",
      "2    a  14.29  14.09  0.9050  5.291  3.337  2.699  4.825  1\n",
      "3    a  13.84  13.94  0.8955  5.324  3.379  2.259  4.805  1\n",
      "4    a  16.14  14.99  0.9034  5.658  3.562  1.355  5.175  1\n",
      "..  ..    ...    ...     ...    ...    ...    ...    ... ..\n",
      "205  d  12.19  13.20  0.8783  5.137  2.981  3.631  4.870  3\n",
      "206  d  11.23  12.88  0.8511  5.140  2.795  4.325  5.003  3\n",
      "207  d  13.20  13.66  0.8883  5.236  3.232  8.315  5.056  3\n",
      "208  d  11.84  13.21  0.8521  5.175  2.836  3.598  5.044  3\n",
      "209  d  12.30  13.34  0.8684  5.243  2.974  5.637  5.063  3\n",
      "\n",
      "[210 rows x 9 columns]\n",
      "            0      1      2       3      4      5      6      7  8\n",
      "0   -0.040990  15.26  14.84  0.8710  5.763  3.312  2.221  5.220  1\n",
      "1   -0.040990  14.88  14.57  0.8811  5.554  3.333  1.018  4.956  1\n",
      "2   -0.040990  14.29  14.09  0.9050  5.291  3.337  2.699  4.825  1\n",
      "3   -0.040990  13.84  13.94  0.8955  5.324  3.379  2.259  4.805  1\n",
      "4   -0.040990  16.14  14.99  0.9034  5.658  3.562  1.355  5.175  1\n",
      "..        ...    ...    ...     ...    ...    ...    ...    ... ..\n",
      "205 -2.985037  12.19  13.20  0.8783  5.137  2.981  3.631  4.870  3\n",
      "206 -2.985037  11.23  12.88  0.8511  5.140  2.795  4.325  5.003  3\n",
      "207 -2.985037  13.20  13.66  0.8883  5.236  3.232  8.315  5.056  3\n",
      "208 -2.985037  11.84  13.21  0.8521  5.175  2.836  3.598  5.044  3\n",
      "209 -2.985037  12.30  13.34  0.8684  5.243  2.974  5.637  5.063  3\n",
      "\n",
      "[210 rows x 9 columns]\n"
     ]
    }
   ],
   "source": [
    "print(D)\n",
    "print(D_new)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     0      1      2       3      4      5      6      7  8\n",
      "0    0  15.26  14.84  0.8710  5.763  3.312  2.221  5.220  1\n",
      "1    0  14.88  14.57  0.8811  5.554  3.333  1.018  4.956  1\n",
      "2    0  14.29  14.09  0.9050  5.291  3.337  2.699  4.825  1\n",
      "3    0  13.84  13.94  0.8955  5.324  3.379  2.259  4.805  1\n",
      "4    0  16.14  14.99  0.9034  5.658  3.562  1.355  5.175  1\n",
      "..  ..    ...    ...     ...    ...    ...    ...    ... ..\n",
      "205  3  12.19  13.20  0.8783  5.137  2.981  3.631  4.870  3\n",
      "206  3  11.23  12.88  0.8511  5.140  2.795  4.325  5.003  3\n",
      "207  3  13.20  13.66  0.8883  5.236  3.232  8.315  5.056  3\n",
      "208  3  11.84  13.21  0.8521  5.175  2.836  3.598  5.044  3\n",
      "209  3  12.30  13.34  0.8684  5.243  2.974  5.637  5.063  3\n",
      "\n",
      "[210 rows x 9 columns]\n",
      "            0      1      2       3      4      5      6      7  8\n",
      "0   -0.040990  15.26  14.84  0.8710  5.763  3.312  2.221  5.220  1\n",
      "1   -0.040990  14.88  14.57  0.8811  5.554  3.333  1.018  4.956  1\n",
      "2   -0.040990  14.29  14.09  0.9050  5.291  3.337  2.699  4.825  1\n",
      "3   -0.040990  13.84  13.94  0.8955  5.324  3.379  2.259  4.805  1\n",
      "4   -0.040990  16.14  14.99  0.9034  5.658  3.562  1.355  5.175  1\n",
      "..        ...    ...    ...     ...    ...    ...    ...    ... ..\n",
      "205 -2.985037  12.19  13.20  0.8783  5.137  2.981  3.631  4.870  3\n",
      "206 -2.985037  11.23  12.88  0.8511  5.140  2.795  4.325  5.003  3\n",
      "207 -2.985037  13.20  13.66  0.8883  5.236  3.232  8.315  5.056  3\n",
      "208 -2.985037  11.84  13.21  0.8521  5.175  2.836  3.598  5.044  3\n",
      "209 -2.985037  12.30  13.34  0.8684  5.243  2.974  5.637  5.063  3\n",
      "\n",
      "[210 rows x 9 columns]\n"
     ]
    }
   ],
   "source": [
    "# labelled data using sklearn ONLY applied to non-numerical data\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "le = LabelEncoder()\n",
    "for j in range(D.shape[1]):\n",
    "    if(D[j].dtype == 'object'):\n",
    "        D[j] = le.fit_transform(D[j])\n",
    "\n",
    "print(D)\n",
    "print(D_new)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[-1.3822179   0.14209777  0.21546244 ... -0.98615174 -0.38357742\n",
      "  -1.22474487]\n",
      " [-1.3822179   0.01118803  0.00822376 ... -1.7881662  -0.92201349\n",
      "  -1.22474487]\n",
      " [-1.3822179  -0.19206658 -0.36020056 ... -0.66747933 -1.18919199\n",
      "  -1.22474487]\n",
      " ...\n",
      " [ 1.25656172 -0.56757084 -0.69024735 ...  3.07658816 -0.71806043\n",
      "   1.22474487]\n",
      " [ 1.25656172 -1.03608992 -1.03564515 ... -0.0681352  -0.7425348\n",
      "   1.22474487]\n",
      " [ 1.25656172 -0.87762023 -0.93586356 ...  1.29122264 -0.70378372\n",
      "   1.22474487]]\n",
      "[[ 0.09682163  0.14209777  0.21546244 ... -0.98615174 -0.38357742\n",
      "  -1.22474487]\n",
      " [ 0.09682163  0.01118803  0.00822376 ... -1.7881662  -0.92201349\n",
      "  -1.22474487]\n",
      " [ 0.09682163 -0.19206658 -0.36020056 ... -0.66747933 -1.18919199\n",
      "  -1.22474487]\n",
      " ...\n",
      " [-1.43678001 -0.56757084 -0.69024735 ...  3.07658816 -0.71806043\n",
      "   1.22474487]\n",
      " [-1.43678001 -1.03608992 -1.03564515 ... -0.0681352  -0.7425348\n",
      "   1.22474487]\n",
      " [-1.43678001 -0.87762023 -0.93586356 ...  1.29122264 -0.70378372\n",
      "   1.22474487]]\n"
     ]
    }
   ],
   "source": [
    "# normalize data using standard scaler\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "UFT_scaler = StandardScaler()\n",
    "UFT_scaled = UFT_scaler.fit_transform(D_new)\n",
    "# print(UFT_scaled)\n",
    "normal_scaler = StandardScaler()\n",
    "normal_scaled = normal_scaler.fit_transform(D)\n",
    "# print(normal_scaled)\n",
    "\n",
    "print(normal_scaled)\n",
    "print(UFT_scaled)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = D.iloc[:, :8].values\n",
    "y = D.iloc[:, 8].values\n",
    "\n",
    "X_uft = D_new.iloc[:, :8].values\n",
    "y_uft = D_new.iloc[:, 8].values\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.79934674 0.81186677 0.79348017 0.29560977 0.64809541 0.73347168\n",
      " 0.18096982 0.58019572]\n",
      "[0.81096879 0.8080119  0.79232032 0.29114068 0.64766457 0.72897432\n",
      " 0.18142334 0.58450016]\n"
     ]
    }
   ],
   "source": [
    "# plot the graph\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.feature_selection import mutual_info_classif as MIC\n",
    "\n",
    "\n",
    "mi_score = MIC(X,y)\n",
    "mi_score_uft = MIC(X_uft,y_uft)\n",
    "\n",
    "print(mi_score)\n",
    "print(mi_score_uft)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KMeans(n_clusters=3, random_state=0)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "kmeans_uft = KMeans(n_clusters=3, random_state=0)\n",
    "kmeans_uft.fit(D_new)\n",
    "kmeans_normal = KMeans(n_clusters=3, random_state=0)\n",
    "kmeans_normal.fit(D)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot data\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# initialize plot 1\n",
    "plt.figure(1)\n",
    "plt.scatter(normal_scaled[:, 1], normal_scaled[:, 2], c=kmeans_normal.labels_, cmap='viridis')\n",
    "plt.xlabel('sepal length')\n",
    "plt.ylabel('sepal width')\n",
    "plt.title('Normalization Only K-means clustering')\n",
    "\n",
    "# initialize plot 2\n",
    "plt.figure(2)\n",
    "plt.scatter(UFT_scaled[:, 1], UFT_scaled[:, 2], c=kmeans_uft.labels_, cmap='viridis')\n",
    "plt.xlabel('sepal length')\n",
    "plt.ylabel('sepal width')\n",
    "plt.title('UFT + Normalization K-means clustering')\n",
    "\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.10.6 64-bit (windows store)",
   "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": "3.10.6"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "ff7573f1831cb99fcd99b3a81494c967ce7bb90e5d2619f47f81c5d8e958e1df"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}