Chapter05/Exercise5.04/Exercise5.04.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exercise 5.04: Breast Cancer Diagnosis Classification using Logistic Regression (with Answers)\n",
    "\n",
    "In this activity we will be using the Breast Cancer dataset [https://archive.ics.uci.edu/ml/datasets/Breast+Cancer+Wisconsin+(Diagnostic)]( https://archive.ics.uci.edu/ml/datasets/Breast+Cancer+Wisconsin+(Diagnostic) ) available under the [UCI Machine Learning Repository] (https://archive.ics.uci.edu/ml/index.php).  The dataset contains characteristics of the cell nuclei present in the digitized image of a fine needle aspirate (FNA) of a breast mass, with the labels _malignant_ and _benign_ for each cell nucleus. Throughout this activity we will use the measurements provided in the dataset to classify between malignant and benign cells.\n",
    "\n",
    "## Import the Required Packages\n",
    "For this exercise we will require the Pandas package for loading the data, the matplotlib package for plotting as well as scikit-learn for creating the Logistic Regression model, doing some feature selection as well as model selection.  Import all of the required packages and relevant modules for these tasks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.feature_selection import SelectKBest\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load the Data\n",
    "Load the Breast Cancer Diagnosis dataset using Pandas and examine the first 5 rows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": true
   },
   "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>mean radius</th>\n",
       "      <th>mean texture</th>\n",
       "      <th>mean perimeter</th>\n",
       "      <th>mean area</th>\n",
       "      <th>mean smoothness</th>\n",
       "      <th>mean compactness</th>\n",
       "      <th>mean concavity</th>\n",
       "      <th>mean concave points</th>\n",
       "      <th>mean symmetry</th>\n",
       "      <th>mean fractal dimension</th>\n",
       "      <th>...</th>\n",
       "      <th>worst texture</th>\n",
       "      <th>worst perimeter</th>\n",
       "      <th>worst area</th>\n",
       "      <th>worst smoothness</th>\n",
       "      <th>worst compactness</th>\n",
       "      <th>worst concavity</th>\n",
       "      <th>worst concave points</th>\n",
       "      <th>worst symmetry</th>\n",
       "      <th>worst fractal dimension</th>\n",
       "      <th>diagnosis</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>17.99</td>\n",
       "      <td>10.38</td>\n",
       "      <td>122.80</td>\n",
       "      <td>1001.0</td>\n",
       "      <td>0.11840</td>\n",
       "      <td>0.27760</td>\n",
       "      <td>0.3001</td>\n",
       "      <td>0.14710</td>\n",
       "      <td>0.2419</td>\n",
       "      <td>0.07871</td>\n",
       "      <td>...</td>\n",
       "      <td>17.33</td>\n",
       "      <td>184.60</td>\n",
       "      <td>2019.0</td>\n",
       "      <td>0.1622</td>\n",
       "      <td>0.6656</td>\n",
       "      <td>0.7119</td>\n",
       "      <td>0.2654</td>\n",
       "      <td>0.4601</td>\n",
       "      <td>0.11890</td>\n",
       "      <td>malignant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>20.57</td>\n",
       "      <td>17.77</td>\n",
       "      <td>132.90</td>\n",
       "      <td>1326.0</td>\n",
       "      <td>0.08474</td>\n",
       "      <td>0.07864</td>\n",
       "      <td>0.0869</td>\n",
       "      <td>0.07017</td>\n",
       "      <td>0.1812</td>\n",
       "      <td>0.05667</td>\n",
       "      <td>...</td>\n",
       "      <td>23.41</td>\n",
       "      <td>158.80</td>\n",
       "      <td>1956.0</td>\n",
       "      <td>0.1238</td>\n",
       "      <td>0.1866</td>\n",
       "      <td>0.2416</td>\n",
       "      <td>0.1860</td>\n",
       "      <td>0.2750</td>\n",
       "      <td>0.08902</td>\n",
       "      <td>malignant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>19.69</td>\n",
       "      <td>21.25</td>\n",
       "      <td>130.00</td>\n",
       "      <td>1203.0</td>\n",
       "      <td>0.10960</td>\n",
       "      <td>0.15990</td>\n",
       "      <td>0.1974</td>\n",
       "      <td>0.12790</td>\n",
       "      <td>0.2069</td>\n",
       "      <td>0.05999</td>\n",
       "      <td>...</td>\n",
       "      <td>25.53</td>\n",
       "      <td>152.50</td>\n",
       "      <td>1709.0</td>\n",
       "      <td>0.1444</td>\n",
       "      <td>0.4245</td>\n",
       "      <td>0.4504</td>\n",
       "      <td>0.2430</td>\n",
       "      <td>0.3613</td>\n",
       "      <td>0.08758</td>\n",
       "      <td>malignant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>11.42</td>\n",
       "      <td>20.38</td>\n",
       "      <td>77.58</td>\n",
       "      <td>386.1</td>\n",
       "      <td>0.14250</td>\n",
       "      <td>0.28390</td>\n",
       "      <td>0.2414</td>\n",
       "      <td>0.10520</td>\n",
       "      <td>0.2597</td>\n",
       "      <td>0.09744</td>\n",
       "      <td>...</td>\n",
       "      <td>26.50</td>\n",
       "      <td>98.87</td>\n",
       "      <td>567.7</td>\n",
       "      <td>0.2098</td>\n",
       "      <td>0.8663</td>\n",
       "      <td>0.6869</td>\n",
       "      <td>0.2575</td>\n",
       "      <td>0.6638</td>\n",
       "      <td>0.17300</td>\n",
       "      <td>malignant</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>20.29</td>\n",
       "      <td>14.34</td>\n",
       "      <td>135.10</td>\n",
       "      <td>1297.0</td>\n",
       "      <td>0.10030</td>\n",
       "      <td>0.13280</td>\n",
       "      <td>0.1980</td>\n",
       "      <td>0.10430</td>\n",
       "      <td>0.1809</td>\n",
       "      <td>0.05883</td>\n",
       "      <td>...</td>\n",
       "      <td>16.67</td>\n",
       "      <td>152.20</td>\n",
       "      <td>1575.0</td>\n",
       "      <td>0.1374</td>\n",
       "      <td>0.2050</td>\n",
       "      <td>0.4000</td>\n",
       "      <td>0.1625</td>\n",
       "      <td>0.2364</td>\n",
       "      <td>0.07678</td>\n",
       "      <td>malignant</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 31 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   mean radius  mean texture  mean perimeter  mean area  mean smoothness  \\\n",
       "0        17.99         10.38          122.80     1001.0          0.11840   \n",
       "1        20.57         17.77          132.90     1326.0          0.08474   \n",
       "2        19.69         21.25          130.00     1203.0          0.10960   \n",
       "3        11.42         20.38           77.58      386.1          0.14250   \n",
       "4        20.29         14.34          135.10     1297.0          0.10030   \n",
       "\n",
       "   mean compactness  mean concavity  mean concave points  mean symmetry  \\\n",
       "0           0.27760          0.3001              0.14710         0.2419   \n",
       "1           0.07864          0.0869              0.07017         0.1812   \n",
       "2           0.15990          0.1974              0.12790         0.2069   \n",
       "3           0.28390          0.2414              0.10520         0.2597   \n",
       "4           0.13280          0.1980              0.10430         0.1809   \n",
       "\n",
       "   mean fractal dimension  ...  worst texture  worst perimeter  worst area  \\\n",
       "0                 0.07871  ...          17.33           184.60      2019.0   \n",
       "1                 0.05667  ...          23.41           158.80      1956.0   \n",
       "2                 0.05999  ...          25.53           152.50      1709.0   \n",
       "3                 0.09744  ...          26.50            98.87       567.7   \n",
       "4                 0.05883  ...          16.67           152.20      1575.0   \n",
       "\n",
       "   worst smoothness  worst compactness  worst concavity  worst concave points  \\\n",
       "0            0.1622             0.6656           0.7119                0.2654   \n",
       "1            0.1238             0.1866           0.2416                0.1860   \n",
       "2            0.1444             0.4245           0.4504                0.2430   \n",
       "3            0.2098             0.8663           0.6869                0.2575   \n",
       "4            0.1374             0.2050           0.4000                0.1625   \n",
       "\n",
       "   worst symmetry  worst fractal dimension  diagnosis  \n",
       "0          0.4601                  0.11890  malignant  \n",
       "1          0.2750                  0.08902  malignant  \n",
       "2          0.3613                  0.08758  malignant  \n",
       "3          0.6638                  0.17300  malignant  \n",
       "4          0.2364                  0.07678  malignant  \n",
       "\n",
       "[5 rows x 31 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('../Datasets/breast-cancer-data.csv')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Dissect the data into input (X) and output (y) variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "X, y = df[[c for c in df.columns if c != 'diagnosis']], df.diagnosis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Engineering\n",
    "We need to select the most appropriate features that will provide the most powerful classification model.  We will use scikit-learn's select K best features sub-module under its feature selection module. Basically it examines the power of each feature against the target output based on a scoring function. You may read about the details here : https://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.SelectKBest.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['worst perimeter', 'worst concave points']\n"
     ]
    }
   ],
   "source": [
    "# restricting to 2 best features so that we can visualize them on a plot\n",
    "skb_model = SelectKBest(k=2)\n",
    "X_new = skb_model.fit_transform(X, y)\n",
    "\n",
    "# get the k - best column names\n",
    "mask = skb_model.get_support() #list of booleans\n",
    "selected_features = [] # The list of your K best features\n",
    "\n",
    "for bool, feature in zip(mask, df.columns):\n",
    "    if bool:\n",
    "        selected_features.append(feature)\n",
    "print(selected_features)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let us visualize how these 2 most important features correlate with the target (diagnosis) and how well do they separate the 2 classes of diagnosis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "markers = {\n",
    "    'benign': {'marker': 'o'},\n",
    "    'malignant': {'marker': 'x'},\n",
    "}\n",
    "plt.figure(figsize=(10, 7))\n",
    "for name, group in df.groupby('diagnosis'):\n",
    "    plt.scatter(group[selected_features[0]], group[selected_features[1]], \n",
    "                label=name,\n",
    "                marker=markers[name]['marker'],\n",
    "               )\n",
    "    \n",
    "plt.title(f'Diagnosis Classification {selected_features[0]} vs {selected_features[1]}');\n",
    "plt.xlabel(selected_features[0]);\n",
    "plt.ylabel(selected_features[1]);\n",
    "plt.legend();\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Constructing the Logistic Regression Model\n",
    "Before we can construct the model we must first convert the dignosis values into labels that can be used within the model.  Replace:\n",
    "\n",
    "1. The diagnosis string *benign* with the value 0\n",
    "2. The diagnosis string *malignant* with the value 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "diagnoses = [\n",
    "    'benign', # 0\n",
    "    'malignant', # 1\n",
    "]\n",
    "output = [diagnoses.index(diag) for diag in df.diagnosis]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Also, in order to impartially evaluate the model, we should split the training dataset into a training and a validation set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_X, valid_X, train_y, valid_y = train_test_split(df[selected_features], output, \n",
    "                                                      test_size=0.2, random_state=123)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create the model using the *selected_features* and the assigned *diagnosis* labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
       "                   multi_class='warn', n_jobs=None, penalty='l2',\n",
       "                   random_state=None, solver='liblinear', tol=0.0001, verbose=0,\n",
       "                   warm_start=False)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = LogisticRegression(solver='liblinear')\n",
    "model.fit(train_X, train_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Compute the accuracy of the model against the validation set:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9385964912280702"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.score(valid_X, valid_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Construct another model using a random choice of *selected_features* and compare the performance:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8859649122807017"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "selected_features = [\n",
    "    'mean radius', # List features here\n",
    "    'mean texture',\n",
    "    'compactness error'\n",
    "]\n",
    "\n",
    "train_X, valid_X, train_y, valid_y = train_test_split(df[selected_features], output, \n",
    "                                                      test_size=0.2, random_state=123)\n",
    "\n",
    "model = LogisticRegression(solver='liblinear')\n",
    "model.fit(train_X, train_y)\n",
    "\n",
    "model.score(valid_X, valid_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This reduced accuracy shows that indeed using the 2 most important features renders a more powerful model than using 3 randomly chosen features."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Construct another model using all available information and compare the performance:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9824561403508771"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "selected_features = [\n",
    "    feat for feat in df.columns if feat != 'diagnosis' # List features here\n",
    "]\n",
    "\n",
    "train_X, valid_X, train_y, valid_y = train_test_split(df[selected_features], output, \n",
    "                                                      test_size=0.2, random_state=123)\n",
    "\n",
    "model = LogisticRegression(solver='liblinear')\n",
    "model.fit(train_X, train_y)\n",
    "\n",
    "model.score(valid_X, valid_y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This improvement in performance by using all features shows that even the features that are not among the most important features do still play a role in improving model performance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}