notebooks/13_CostSensitiveClassification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 13 -  Cost-Sensitive Classification\n",
    "\n",
    "by [Alejandro Correa Bahnsen](albahnsen.com/)\n",
    "\n",
    "version 0.2, May 2016\n",
    "\n",
    "## Part of the class [Machine Learning for Risk Management](https://github.com/albahnsen/ML_RiskManagement)\n",
    "\n",
    "This notebook is licensed under a [Creative Commons Attribution-ShareAlike 3.0 Unported License](http://creativecommons.org/licenses/by-sa/3.0/deed.en_US)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div>\n",
    "<img img class=\"logo\" src=\"https://raw.githubusercontent.com/albahnsen/CostSensitiveClassification/master/logo.png\" style=\"width: 300px;\" align=\"right\">\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Install costcla\n",
    "\n",
    "```\n",
    "pip install costcla\n",
    "```\n",
    "\n",
    "see https://github.com/albahnsen/CostSensitiveClassification"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "  ## Credit Scoring\n",
    "  \n",
    "  In order to mitigate the impact of credit risk and make more objective and accurate decisions, \n",
    "  financial institutions use credit scores to predict and control their losses.\n",
    "  The objective in credit scoring is to classify which potential customers are likely to default a \n",
    "  contracted financial obligation based on the customer's past financial experience, and with that \n",
    "  information decide whether to approve or decline a loan [1]. This tool has \n",
    "  become a standard practice among financial institutions around the world in order to predict \n",
    "  and control their loans portfolios. When constructing credit scores, it is a common practice to \n",
    "  use standard cost-insensitive binary classification algorithms such as logistic regression, \n",
    "  neural networks, discriminant analysis, genetic programing, decision trees, among \n",
    "  others [2,3]. \n",
    "  \n",
    "  Formally, a credit score is a statistical model that allows the estimation of the probability \n",
    "  $\\hat p_i=P(y_i=1|\\mathbf{x}_i)$ of a customer $i$ defaulting a contracted debt. Additionally, \n",
    "since the objective of credit scoring is to estimate a classifier $c_i$ to decide whether or not to grant a \n",
    "loan to a customer $i$, a threshold $t$ is defined such that if $\\hat p_i < t $, then the loan is \n",
    "  granted, i.e., $c_i(t)=0$, and denied otherwise, i.e., $c_i(t)=1$."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Example: Kaggle Credit Competition\n",
    "\n",
    "Improve on the state of the art in credit scoring by predicting the probability that somebody will experience financial distress in the next two years.\n",
    "\n",
    "https://www.kaggle.com/c/GiveMeSomeCredit\n",
    "\n",
    "### Load dataset and show basic statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['cost_mat', 'name', 'data', 'target', 'target_names', 'feature_names', 'DESCR'])\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from costcla.datasets import load_creditscoring1\n",
    "data = load_creditscoring1()\n",
    "\n",
    "# Elements of the data file\n",
    "print(data.keys())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>RevolvingUtilizationOfUnsecuredLines</th>\n",
       "      <th>age</th>\n",
       "      <th>NumberOfTime30-59DaysPastDueNotWorse</th>\n",
       "      <th>DebtRatio</th>\n",
       "      <th>MonthlyIncome</th>\n",
       "      <th>NumberOfOpenCreditLinesAndLoans</th>\n",
       "      <th>NumberOfTimes90DaysLate</th>\n",
       "      <th>NumberRealEstateLoansOrLines</th>\n",
       "      <th>NumberOfTime60-89DaysPastDueNotWorse</th>\n",
       "      <th>NumberOfDependents</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.766127</td>\n",
       "      <td>45.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.802982</td>\n",
       "      <td>9120.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.957151</td>\n",
       "      <td>40.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.121876</td>\n",
       "      <td>2600.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.658180</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.085113</td>\n",
       "      <td>3042.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.233810</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.036050</td>\n",
       "      <td>3300.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.907239</td>\n",
       "      <td>49.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.024926</td>\n",
       "      <td>63588.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   RevolvingUtilizationOfUnsecuredLines   age  \\\n",
       "0                              0.766127  45.0   \n",
       "1                              0.957151  40.0   \n",
       "2                              0.658180  38.0   \n",
       "3                              0.233810  30.0   \n",
       "4                              0.907239  49.0   \n",
       "\n",
       "   NumberOfTime30-59DaysPastDueNotWorse  DebtRatio  MonthlyIncome  \\\n",
       "0                                   2.0   0.802982         9120.0   \n",
       "1                                   0.0   0.121876         2600.0   \n",
       "2                                   1.0   0.085113         3042.0   \n",
       "3                                   0.0   0.036050         3300.0   \n",
       "4                                   1.0   0.024926        63588.0   \n",
       "\n",
       "   NumberOfOpenCreditLinesAndLoans  NumberOfTimes90DaysLate  \\\n",
       "0                             13.0                      0.0   \n",
       "1                              4.0                      0.0   \n",
       "2                              2.0                      1.0   \n",
       "3                              5.0                      0.0   \n",
       "4                              7.0                      0.0   \n",
       "\n",
       "   NumberRealEstateLoansOrLines  NumberOfTime60-89DaysPastDueNotWorse  \\\n",
       "0                           6.0                                   0.0   \n",
       "1                           0.0                                   0.0   \n",
       "2                           0.0                                   0.0   \n",
       "3                           0.0                                   0.0   \n",
       "4                           1.0                                   0.0   \n",
       "\n",
       "   NumberOfDependents  \n",
       "0                 2.0  \n",
       "1                 1.0  \n",
       "2                 0.0  \n",
       "3                 0.0  \n",
       "4                 0.0  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(data.data, columns=data.feature_names).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                           Frequency  Percentage\n",
      "Negative (Good Customers)     105299    0.932551\n",
      "Positive (Bad Customers)        7616    0.067449\n"
     ]
    }
   ],
   "source": [
    "# Class label\n",
    "target = pd.DataFrame(pd.Series(data.target).value_counts(), columns=('Frequency',))\n",
    "target['Percentage'] = target['Frequency'] / target['Frequency'].sum()\n",
    "target.index = ['Negative (Good Customers)', 'Positive (Bad Customers)']\n",
    "print(target)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Full description of the dataset\n",
    "# print data.DESCR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Features</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>RevolvingUtilizationOfUnsecuredLines</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>age</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>NumberOfTime30-59DaysPastDueNotWorse</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DebtRatio</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>MonthlyIncome</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>NumberOfOpenCreditLinesAndLoans</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>NumberOfTimes90DaysLate</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>NumberRealEstateLoansOrLines</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>NumberOfTime60-89DaysPastDueNotWorse</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>NumberOfDependents</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               Features\n",
       "0  RevolvingUtilizationOfUnsecuredLines\n",
       "1                                   age\n",
       "2  NumberOfTime30-59DaysPastDueNotWorse\n",
       "3                             DebtRatio\n",
       "4                         MonthlyIncome\n",
       "5       NumberOfOpenCreditLinesAndLoans\n",
       "6               NumberOfTimes90DaysLate\n",
       "7          NumberRealEstateLoansOrLines\n",
       "8  NumberOfTime60-89DaysPastDueNotWorse\n",
       "9                    NumberOfDependents"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Number of features\n",
    "pd.DataFrame(data.feature_names, columns=('Features',))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Credit scoring as a standard classification problem\n",
    "\n",
    "Using three classifiers, a model is learned to classify customers in good and bad"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# Load classifiers and split dataset in training and testing\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.cross_validation import train_test_split\n",
    "X_train, X_test, y_train, y_test, cost_mat_train, cost_mat_test = \\\n",
    "train_test_split(data.data, data.target, data.cost_mat)\n",
    "\n",
    "# Fit the classifiers using the training dataset\n",
    "classifiers = {\"RF\": {\"f\": RandomForestClassifier()},\n",
    "               \"DT\": {\"f\": DecisionTreeClassifier()},\n",
    "               \"LR\": {\"f\": LogisticRegression()}}\n",
    "\n",
    "for model in classifiers.keys():\n",
    "    # Fit\n",
    "    classifiers[model][\"f\"].fit(X_train, y_train)\n",
    "    # Predict\n",
    "    classifiers[model][\"c\"] = classifiers[model][\"f\"].predict(X_test)\n",
    "    classifiers[model][\"p\"] = classifiers[model][\"f\"].predict_proba(X_test)\n",
    "    classifiers[model][\"p_train\"] = classifiers[model][\"f\"].predict_proba(X_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "  After the classifier $c_i$ is estimated, there is a need to evaluate its performance. In \n",
    "  practice, many statistical evaluation measures are used to assess the performance of a credit \n",
    "  scoring model. Measures such as the area under the  receiver operating characteristic curve (AUC),\n",
    "  Brier score, Kolmogorov-Smirnoff (K-S) statistic,  $F_1$-Score, and misclassification are among \n",
    "  the most common [4]. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "          f1       pre       acc       rec\n",
      "DT  0.258659  0.245158  0.894612  0.273734\n",
      "LR  0.021739  0.583333  0.933048  0.011076\n",
      "RF  0.238577  0.511265  0.933296  0.155591\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the performance\n",
    "from sklearn.metrics import f1_score, precision_score, recall_score, accuracy_score\n",
    "measures = {\"f1\": f1_score, \"pre\": precision_score, \n",
    "            \"rec\": recall_score, \"acc\": accuracy_score}\n",
    "results = pd.DataFrame(columns=measures.keys())\n",
    "\n",
    "# Evaluate each model in classifiers\n",
    "for model in classifiers.keys():\n",
    "    results.loc[model] = [measures[measure](y_test, classifiers[model][\"c\"]) for measure in measures.keys()]\n",
    "\n",
    "print(results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAApoAAAE8CAYAAABgoRGRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nVs9Y0V917t8AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f93c4e349e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the results\n",
    "%matplotlib inline\n",
    "from IPython.core.pylabtools import figsize\n",
    "import matplotlib.pyplot as plt\n",
    "plt.style.use('ggplot')\n",
    "\n",
    "figsize(10, 5)\n",
    "ax = plt.subplot(111)\n",
    "\n",
    "ind = np.arange(results.shape[0])\n",
    "width = 0.2\n",
    "l = ax.plot(ind, results, \"-o\")\n",
    "plt.legend(iter(l), results.columns.tolist(), loc='center left', bbox_to_anchor=(1, 0.5))\n",
    "ax.set_xlim([-0.25, ind[-1]+.25])\n",
    "ax.set_xticks(ind)\n",
    "ax.set_xticklabels(results.index)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "  Nevertheless, none of these measures takes into account the \n",
    "  business and economical realities that take place in credit scoring. Costs that the financial \n",
    "  institution had incurred to acquire customers, or the expected profit due to a particular client, \n",
    "  are not considered in the evaluation of the different models. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Financial Evaluation of a Credit Scorecard\n",
    "\n",
    "  Typically, a credit risk model is evaluated using standard cost-insensitive measures.\n",
    "  However, in practice, the cost associated with approving \n",
    "  what is known as a bad customer, i.e., a customer who default his credit loan, is quite \n",
    "  different from the cost associated with declining a good customer,  i.e., a customer who \n",
    "  successfully repay his credit loan. Furthermore, the costs are not constant among customers. \n",
    "  This is because loans have different credit line amounts, terms, and even interest rates. Some \n",
    "  authors have proposed methods that include the misclassification costs in the credit scoring \n",
    "  context [4,5,6,7].\n",
    "  \n",
    "  In order to take into account the varying costs that each example carries, we proposed in \n",
    "  [8], a cost matrix with example-dependent misclassification costs as \n",
    "  given in the following table.\n",
    "  \n",
    "  \n",
    "|  \t| Actual Positive ($y_i=1$)  \t|  Actual Negative \t($y_i=0$)|\n",
    "|---\t|:-:\t|:-:\t|\n",
    "|   Predicted Positive ($c_i=1$)\t|   $C_{TP_i}=0$\t|  $C_{FP_i}=r_i+C^a_{FP}$ \t|\n",
    "|  Predicted Negative  ($c_i=0$) \t|   $C_{FN_i}=Cl_i \\cdot L_{gd}$\t| $C_{TN_i}=0$\t|\n",
    "  \n",
    "  First, we assume that the costs of a correct \n",
    "  classification, $C_{TP_i}$ and $C_{TN_i}$, are zero for every customer $i$. We define $C_{FN_i}$ \n",
    "  to be the losses if the customer $i$ defaults to be proportional to his credit line $Cl_i$. We \n",
    "  define the cost of a false positive per customer $C_{FP_i}$ as the sum of two real financial \n",
    "  costs $r_i$ and $C^a_{FP}$, where $r_i$ is the loss in profit by rejecting what would have been a \n",
    "  good customer. \n",
    "  \n",
    "  The profit per customer $r_i$ is calculated as the present value of the difference between the \n",
    "  financial institution gains and expenses, given the credit line $Cl_i$, the term $l_i$ and the \n",
    "  financial institution lending rate $int_{r_i}$ for customer $i$, and the financial institution \n",
    "  of cost funds $int_{cf}$.\n",
    "\n",
    "  $$  r_i= PV(A(Cl_i,int_{r_i},l_i),int_{cf},l_i)-Cl_i,$$\n",
    "  \n",
    "  with $A$ being the customer monthly payment and $PV$ the present value of the monthly payments,\n",
    "  which are calculated using the time value of money equations [9],\n",
    " \n",
    " $$ A(Cl_i,int_{r_i},l_i) =  Cl_i \\frac{int_{r_i}(1+int_{r_i})^{l_i}}{(1+int_{r_i})^{l_i}-1},$$\n",
    " \n",
    " $$ PV(A,int_{cf},l_i) = \\frac{A}{int_{cf}} \\left(1-\\frac{1}{(1+int_{cf})^{l_i}} \\right).$$\n",
    "      \n",
    "  The second term $C^a_{FP}$, is related to the assumption that the financial institution will not \n",
    "  keep the money of the declined customer idle. It will instead give a loan to an alternative \n",
    "  customer [10]. Since no further information is known about the alternative customer, \n",
    "  it is assumed to have an average credit line $\\overline{Cl}$ and an average profit $\\overline{r}$.\n",
    "  Given that, \n",
    "  \n",
    "  $$  C^a_{FP}=- \\overline{r} \\cdot \\pi_0+\\overline{Cl}\\cdot L_{gd} \\cdot \\pi_1,$$\n",
    "\n",
    "  in other words minus the profit of an average alternative customer plus the expected loss, \n",
    "  taking into account that the alternative customer will pay his debt with a probability equal to \n",
    "  the prior negative rate, and similarly will default with probability equal to the prior positive \n",
    "  rate.\n",
    "  \n",
    "  One key parameter of our model is the credit limit. There exists several strategies to calculate \n",
    "  the $Cl_i$ depending on the type of loans, the state of the economy, the current portfolio, \n",
    "  among others [1,9]. Nevertheless, given the lack of information \n",
    "  regarding the specific business environments of the considered datasets, we simply define \n",
    "  $Cl_i$ as\n",
    "\n",
    "$$      Cl_i = \\min \\bigg\\{ q \\cdot Inc_i, Cl_{max}, Cl_{max}(debt_i) \\bigg\\},$$\n",
    "  \n",
    "  where $Inc_i$ and $debt_i$ are the monthly income and debt ratio of the customer $i$, \n",
    "  respectively, $q$ is a parameter that defines the maximum $Cl_i$ in times $Inc_i$, and \n",
    "  $Cl_{max}$ the maximum overall credit line. Lastly, the maximum credit line given the current \n",
    "  debt is calculated as the maximum credit limit such that the current debt ratio plus the new \n",
    "  monthly payment does not surpass the customer monthly income. It is calculated as\n",
    " \n",
    " $$  Cl_{max}(debt_i)=PV\\left(Inc_i \\cdot P_{m}(debt_i),int_{r_i},l_i\\right),$$\n",
    "  and\n",
    "  $$ P_{m}(debt_i)=\\min \\left\\{ \\frac{A(q \\cdot Inc_i,int_{r_i},l_i)}{Inc_i},\\left(1-debt_i \\right) \\right\\}.$$\n",
    "  \n",
    "  \n",
    "### Financial savings\n",
    "\n",
    "  Let $\\mathcal{S}$ be a set of $N$ examples $i$, $N=\\vert S \\vert$, where each example is \n",
    "  represented by  the augmented feature vector \n",
    "  $\\mathbf{x}_i^*=[\\mathbf{x}_i, C_{TP_i},C_{FP_i},C_{FN_i},C_{TN_i}]$  \n",
    "  and labeled using the class   label $y_i   \\in \\{0,1\\}$. \n",
    "  A classifier $f$ which generates the   predicted label $c_i$ for each   element $i$ is trained  \n",
    "  using the set $\\mathcal{S}$. Then the cost of   using $f$ on $\\mathcal{S}$ is calculated by\n",
    "  \n",
    "  $$   Cost(f(\\mathcal{S})) = \\sum_{i=1}^N Cost(f(\\mathbf{x}_i^*)),$$\n",
    "  \n",
    "  where\n",
    "  \n",
    " $$   Cost(f(\\mathbf{x}_i^*)) = y_i(c_i C_{TP_i} + (1-c_i)C_{FN_i}) + (1-y_i)(c_i C_{FP_i} + (1-c_i)C_{TN_i}).$$\n",
    "  \n",
    "\n",
    "  However, the total cost may not be easy to interpret. We proposed an approach in [8], where the savings of using an algorithm  are defined as the cost of the algorithm versus the cost of using no algorithm at all.  To do that, the cost of the costless class is defined as \n",
    "  \n",
    "  $$  Cost_l(\\mathcal{S}) = \\min \\{Cost(f_0(\\mathcal{S})), Cost(f_1(\\mathcal{S}))\\},$$\n",
    "  \n",
    "  where \n",
    "  \n",
    "  $$  f_a(\\mathcal{S}) = \\mathbf{a}, \\text{ with } a\\in \\{0,1\\}.$$\n",
    "  \n",
    "\n",
    "  The cost improvement can be expressed as the cost savings as compared with $Cost_l(\\mathcal{S})$. \n",
    "  \n",
    "  $$    Savings(f(\\mathcal{S})) = \\frac{ Cost_l(\\mathcal{S}) - Cost(f(\\mathcal{S}))}   {Cost_l(\\mathcal{S})}.$$\n",
    "  \n",
    "\n",
    "\n",
    "  ### Parameters for the Kaggle Credit Database\n",
    "\n",
    " As this database contain information regarding the features, and more importantly about the income of each example, from which an estimated credit limit $Cl_i$ can be calculated.\n",
    "Since no specific information regarding the datasets is provided, we assume that they belong to  an\n",
    "average Portuguese financial institution. This enabled us to find the different \n",
    "parameters needed to calculate the cost measure. \n",
    "\n",
    "| Parameter \t| Value |\n",
    "|---\t|:-:\t|\n",
    "|Interest rate ($int_r$) | 4.79% |\n",
    "|  Cost of funds ($int_{cf}$) | 2.94% |\n",
    "|  Term ($l$) in months | 24 |\n",
    "|  Loss given default ($L_{gd}$) | 75% |\n",
    "|  Times income ($q$) | 3 |\n",
    "|  Maximum credit line ($Cl_{max}$) | 25,000|\n",
    "\n",
    "In particular, we obtain the average interest rates in Europe during   2013 from the European Central Bank [11]. Additionally, we use a fixed loan term $l$ to two years,\n",
    "considering that in the Kaggle Credit dataset the class was constructed to predict two years of credit behavior.\n",
    "Moreover, we set the loss given default $L_{gd}$ using information from \n",
    "the Basel II standard, $q$ to 3 since it is the average personal loan requests related to monthly income, and the maximum credit limit $Cl_{max}$ to 25,000 Euros.\n",
    "\n",
    "### Calculation of the savings of the models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[  1023.73054104  18750.              0.              0.        ]\n",
      " [   717.25781516   6749.25            0.              0.        ]\n",
      " [   866.65393177  12599.25            0.              0.        ]]\n"
     ]
    }
   ],
   "source": [
    "# The cost matrix is already calculated for the dataset\n",
    "# cost_mat[C_FP,C_FN,C_TP,C_TN]\n",
    "print(data.cost_mat[[10, 17, 50]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "          f1       pre       acc       rec       sav\n",
      "DT  0.258659  0.245158  0.894612  0.273734  0.191818\n",
      "LR  0.021739  0.583333  0.933048  0.011076  0.003708\n",
      "RF  0.238577  0.511265  0.933296  0.155591  0.127245\n"
     ]
    }
   ],
   "source": [
    "# Calculation of the cost and savings\n",
    "from costcla.metrics import savings_score, cost_loss \n",
    "\n",
    "# Evaluate the savings for each model\n",
    "results[\"sav\"] = np.zeros(results.shape[0])\n",
    "for model in classifiers.keys():\n",
    "    results[\"sav\"].loc[model] = savings_score(y_test, classifiers[model][\"c\"], cost_mat_test)\n",
    "\n",
    "# TODO: plot results\n",
    "print(results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArsAAAE8CAYAAADAPnl/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n3sjyaj0AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f93c43bc128>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the results\n",
    "\n",
    "figsize(10, 5)\n",
    "ax = plt.subplot(111)\n",
    "l = ax.plot(ind, results[\"f1\"], \"-o\", label='F1Score')\n",
    "b = ax.bar(ind-0.3, results['sav'], 0.6, label='Savings')\n",
    "plt.legend(loc='center left', bbox_to_anchor=(1, 0.5))\n",
    "ax.set_xlim([-0.5, ind[-1]+.5])\n",
    "ax.set_xticks(ind)\n",
    "ax.set_xticklabels(results.index)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are significant differences in the \n",
    "results when evaluating a model using a  traditional cost-insensitive measure such as the \n",
    "accuracy or F1Score,  than when using the savings, leading to the conclusion of the \n",
    "importance of using the real practical financial costs of each context."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Bayes minimum risk\n",
    "\n",
    "As these methods (RF, LR and DT) are not performing well we then move to use cost-sensitive methods. The first model we used is the Bayes minimum risk model (BMR) [8].\n",
    "As defined in [12], the BMR classifier is a decision model based on quantifying \n",
    "tradeoffs between various decisions using probabilities and the costs that accompany such decisions. \n",
    "This is done in a way that for each example the expected losses are minimized. In  what follows, we \n",
    "consider the probability estimates $\\hat p_i$ as known, regardless of the algorithm used to \n",
    "calculate them.  The risk that accompanies each decision is calculated using the cost matrix described above.\n",
    "In the specific framework of binary classification, the risk of predicting the example $i$ as negative is \n",
    "\n",
    "$$ R(c_i=0|\\mathbf{x}_i)=C_{TN_i}(1-\\hat p_i)+C_{FN_i} \\cdot \\hat p_i, $$\n",
    "and\n",
    "$$ R(c_i=1|\\mathbf{x}_i)=C_{TP_i} \\cdot \\hat p_i + C_{FP_i}(1- \\hat p_i), $$\n",
    "\n",
    "is the risk when predicting the example as positive, where $\\hat p_i$ is the estimated positive \n",
    "probability for example $i$. Subsequently, if \n",
    "\n",
    "$$  R(c_i=0|\\mathbf{x}_i) \\le R(c_i=1|\\mathbf{x}_i), $$\n",
    "\n",
    "then  the example $i$ is classified as negative. This means that the risk associated with the \n",
    "decision $c_i$ is lower than the risk associated with classifying it as positive. \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              f1       pre       acc       rec       sav\n",
      "DT      0.258659  0.245158  0.894612  0.273734  0.191818\n",
      "LR      0.021739  0.583333  0.933048  0.011076  0.003708\n",
      "RF      0.238577  0.511265  0.933296  0.155591  0.127245\n",
      "DT-BMR  0.136897  0.082195  0.653371  0.409283  0.086920\n",
      "LR-BMR  0.169111  0.099707  0.632754  0.556435  0.184102\n",
      "RF-BMR  0.264445  0.166920  0.762337  0.636076  0.402713\n"
     ]
    }
   ],
   "source": [
    "from costcla.models import BayesMinimumRiskClassifier\n",
    "ci_models = list(classifiers.keys())\n",
    "\n",
    "for model in ci_models:\n",
    "    classifiers[model+\"-BMR\"] = {\"f\": BayesMinimumRiskClassifier()}\n",
    "    # Fit\n",
    "    classifiers[model+\"-BMR\"][\"f\"].fit(y_test, classifiers[model][\"p\"])  \n",
    "    # Calibration must be made in a validation set\n",
    "    # Predict\n",
    "    classifiers[model+\"-BMR\"][\"c\"] = classifiers[model+\"-BMR\"][\"f\"].predict(classifiers[model][\"p\"], cost_mat_test)\n",
    "    # Evaluate\n",
    "    results.loc[model+\"-BMR\"] = 0\n",
    "    results.loc[model+\"-BMR\", measures.keys()] = \\\n",
    "    [measures[measure](y_test, classifiers[model+\"-BMR\"][\"c\"]) for measure in measures.keys()]\n",
    "    results[\"sav\"].loc[model+\"-BMR\"] = savings_score(y_test, classifiers[model+\"-BMR\"][\"c\"], cost_mat_test)\n",
    "    \n",
    "print(results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArsAAAE8CAYAAADAPnl/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f93c47512e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the results\n",
    "ind = np.arange(results.shape[0])\n",
    "figsize(10, 5)\n",
    "ax = plt.subplot(111)\n",
    "l = ax.plot(ind, results[\"f1\"], \"-o\", label='F1Score')\n",
    "b = ax.bar(ind-0.3, results['sav'], 0.6, label='Savings')\n",
    "plt.legend(loc='center left', bbox_to_anchor=(1, 0.5))\n",
    "ax.set_xlim([-0.5, ind[-1]+.5])\n",
    "ax.set_xticks(ind)\n",
    "ax.set_xticklabels(results.index)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cost-sensitive decision trees\n",
    "\n",
    "The next algorithm that is evaluated is the cost-sensitive decision trees algorithm [13].\n",
    "\n",
    "Decision trees are one of the most widely used machine learning algorithms [14]. \n",
    "The technique is considered to be white box, in the sense that is easy to interpret, and has a \n",
    "very low computational cost, while maintaining a good performance as compared with more complex \n",
    "techniques [15]. There are two types of decision tree depending on the objective of \n",
    "the model. They work either for classification or regression. \n",
    "\n",
    "- Construction of classification trees\n",
    "\n",
    "Classification trees is one of the most common types of decision tree, in which the objective \n",
    "is to find the $Tree$ that best discriminates between classes. In general the decision tree \n",
    "represents a set of splitting rules organized in levels in a flowchart structure.\n",
    "\n",
    "- Splitting criteria\n",
    "\n",
    "In the $Tree$, each rule is shown as a node, and it is represented as $(\\mathbf{x}^j,l^j)$, meaning \n",
    "that the set $\\mathcal{S}$ is split in two sets $\\mathcal{S}^l$ and $\\mathcal{S}^r$ according to \n",
    "$\\mathbf{x}^j$ and $l^j$:\n",
    "\n",
    "$$ \\mathcal{S}^l = \\{\\mathbf{x}_i^* \\vert \\mathbf{x}_i^* \\in \\mathcal{S} \\wedge x^j_i \\le l^j \\}, \\quad and \\quad\n",
    "  \\mathcal{S}^r = \\{\\mathbf{x}_i^* \\vert \\mathbf{x}_i^* \\in \\mathcal{S} \\wedge x^j_i > l^j \\}, $$\n",
    "\n",
    "where $\\mathbf{x}^j$ is the $j^{th}$ feature represented in the vector \n",
    "$\\mathbf{x}^j=[x_1^j,x_2^j,...,x_N^j]$, and $l^j$ is a value such that $min(\\mathbf{x}^j) \\le l^j < \n",
    "max(\\mathbf{x}^j)$. Moreover, $\\mathcal{S} = \\mathcal{S}^l \\cup \\mathcal{S}^r$.\n",
    "\n",
    "After the training set have been split, the percentage of positives in the different sets is calculated. First, the number of positives in each set is estimated by\n",
    "\n",
    "$$ \\mathcal{S}_1 = \\{\\mathbf{x}_i^* \\vert \\mathbf{x}_i^* \\in \\mathcal{S} \\wedge y_i =1 \\}, $$\n",
    "\n",
    "and the percentage of positives is calculates as $\\pi_1=\\vert \\mathcal{S}_1 \\vert / \\vert \\mathcal{S} \\vert.$\n",
    "\n",
    "Then, the impurity of each leaf is calculated using either a misclassification error, \n",
    "entropy or Gini measures:\n",
    "\n",
    "\n",
    "a) Misclassification: \n",
    "    $I_m(\\pi_1)=1-max(\\pi_1,1-\\pi_1)$ \n",
    "\n",
    "b) Entropy: $I_e(\\pi_1)=-\\pi_1\\log \\pi_1 -(1-\\pi_1) \\log (1-\\pi_1)$ \n",
    "\n",
    "c) Gini: \n",
    "    $I_g(\\pi_1)=2\\pi_1(1-\\pi_1)$\n",
    "\n",
    "Finally the gain of the splitting criteria using the rule $(\\mathbf{x}^j,l^j)$ is calculated as the \n",
    "impurity of $\\mathcal{S}$ minus the weighted impurity of each leaf:\n",
    "\n",
    "$$  Gain(\\mathbf{x}^j,l^j)=I(\\pi_1)-\\frac{\\vert \\mathcal{S}^l \\vert}{\\vert \\mathcal{S} \n",
    "\\vert}I(\\pi^l_1)  -\\frac{\\vert \\mathcal{S}^r \\vert}{\\vert \\mathcal{S} \\vert}I(\\pi^r_1), $$\n",
    "\n",
    "where $I(\\pi_1)$ can be either of the impurity measures $I_e(\\pi_1)$ or $I_g(\\pi_1)$.\n",
    "\n",
    "Subsequently, the gain of all possible splitting rules is calculated. The rule with maximal \n",
    "gain is selected\n",
    "\n",
    "$$  (best_x, best_l) = argmax_{(\\mathbf{x}^j,l^j)}Gain(\\mathbf{x}^j,l^j), $$\n",
    "\n",
    "and the set $\\mathcal{S}$ is split into $\\mathcal{S}^l$ and $\\mathcal{S}^r$ according to that rule. \n",
    "\n",
    "\n",
    "- Tree growing\n",
    "\n",
    "In order to grow a tree typical algorithms use a top-down induction using a greedy search in each\n",
    "iteration [16]. In each iteration, the algorithms evaluates all possible splitting \n",
    "rules and pick the one that maximizes the splitting criteria. After the selection of a splitting \n",
    "rule, each leaf is further selected and it is subdivides into smaller leafs, until one of the \n",
    "stopping criteria is meet. \n",
    "\n",
    "- Cost-sensitive impurity measures\n",
    "\n",
    "Standard impurity measures such as misclassification, entropy or Gini, take into account the \n",
    "distribution of classes of each leaf to evaluate the predictive power of a splitting rule,\n",
    "leading to an impurity measure that is based on minimizing the misclassification rate. However, \n",
    "as has been previously shown [8], minimizing misclassification does not \n",
    "lead to the same results than minimizing cost. Instead, we are interested in measuring how good \n",
    "is a splitting rule in terms of cost not only accuracy. For doing that, we propose a new \n",
    "example-dependent cost based impurity measure that takes into account the cost matrix of each \n",
    "example.\n",
    "\n",
    "We define a new cost-based impurity measure taking into account the costs when all the examples\n",
    "in a leaf are classified both as negative using $f_0$ and positive using $f_1$\n",
    "\n",
    "$$    I_c(\\mathcal{S}) = \\min \\bigg\\{ Cost(f_0(\\mathcal{S})), Cost(f_1(\\mathcal{S})) \\bigg\\}. $$\n",
    "\n",
    "The objective of this measure is to evaluate the lowest expected cost of a splitting rule.\n",
    "Following the same logic, the classification of each set is calculated as the prediction that \n",
    "leads to the lowest cost\n",
    "\n",
    "$$\n",
    "f(\\mathcal{S}) = \n",
    "\\begin{cases}\n",
    "  \\phantom{-}0 \\phantom{-} \\mbox{if} \\phantom{-} Cost(f_0(\\mathcal{S})) \\le \n",
    "Cost(f_1(\\mathcal{S}))\\\\\n",
    "  \\phantom{-}1 \\phantom{-}\\mbox{otherwise}\n",
    "\\end{cases}\n",
    "$$\n",
    "\n",
    "Finally, using the cost-based impurity, the splitting criteria cost based gain of using the \n",
    "splitting rule $(\\mathbf{x}^j,l^j)$ is calculated. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              f1       pre       acc       rec       sav\n",
      "DT      0.258659  0.245158  0.894612  0.273734  0.191818\n",
      "LR      0.021739  0.583333  0.933048  0.011076  0.003708\n",
      "RF      0.238577  0.511265  0.933296  0.155591  0.127245\n",
      "DT-BMR  0.136897  0.082195  0.653371  0.409283  0.086920\n",
      "LR-BMR  0.169111  0.099707  0.632754  0.556435  0.184102\n",
      "RF-BMR  0.264445  0.166920  0.762337  0.636076  0.402713\n",
      "CSDT    0.270222  0.162937  0.712990  0.791139  0.464976\n"
     ]
    }
   ],
   "source": [
    "from costcla.models import CostSensitiveDecisionTreeClassifier\n",
    "\n",
    "classifiers[\"CSDT\"] = {\"f\": CostSensitiveDecisionTreeClassifier()}\n",
    "# Fit\n",
    "classifiers[\"CSDT\"][\"f\"].fit(X_train, y_train, cost_mat_train)\n",
    "# Predict\n",
    "classifiers[\"CSDT\"][\"c\"] = classifiers[\"CSDT\"][\"f\"].predict(X_test)\n",
    "# Evaluate\n",
    "results.loc[\"CSDT\"] = 0\n",
    "results.loc[\"CSDT\", measures.keys()] = \\\n",
    "[measures[measure](y_test, classifiers[\"CSDT\"][\"c\"]) for measure in measures.keys()]\n",
    "results[\"sav\"].loc[\"CSDT\"] = savings_score(y_test, classifiers[\"CSDT\"][\"c\"], cost_mat_test)\n",
    "    \n",
    "print(results)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArUAAAE8CAYAAADe90nMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f93c472de10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the results\n",
    "ind = np.arange(results.shape[0])\n",
    "figsize(10, 5)\n",
    "ax = plt.subplot(111)\n",
    "l = ax.plot(ind, results[\"f1\"], \"-o\", label='F1Score')\n",
    "b = ax.bar(ind-0.3, results['sav'], 0.6, label='Savings')\n",
    "plt.legend(loc='center left', bbox_to_anchor=(1, 0.5))\n",
    "ax.set_xlim([-0.5, ind[-1]+.5])\n",
    "ax.set_xticks(ind)\n",
    "ax.set_xticklabels(results.index)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cost-Sensitive Random Patches\n",
    "\n",
    "Ensemble learning is a widely studied topic in the machine learning community. The main idea behind the ensemble methodology is to combine several individual base classifiers in   order to have a \n",
    "classifier that outperforms each of them [17]. Nowadays,   ensemble methods are  one of the most popular and well studied machine learning techniques [18], and it can be noted that since 2009 all the first-place and   second-place winners of the KDD-Cup (https://www.sigkdd.org/kddcup/)   used  ensemble methods. The core \n",
    "principle in ensemble learning, is to induce random perturbations into  the learning procedure in order to produce several different base classifiers from a single  training set, then combining the \n",
    "base classifiers in order to make the final prediction.  In order to induce the random permutations and therefore create the different base classifiers,   several methods have been proposed, in \n",
    "particular: bagging,   pasting, random forests and random patches [19]. Finally, after  the base   classifiers are trained, they are typically   combined using either   majority voting,  weighted  voting    or  stacking [18].\n",
    "\n",
    "With the objective of creating an ensemble of example-dependent cost-sensitive decision trees, we first create $T$ different random subsamples $\\mathcal{S}_j$ for $j=1,\\dots,T$, of the training  set \n",
    "$\\mathcal{S}$, and train a $CSDT$ algorithm on each one. In particular we create the different [19]. In the random patches algorithm, base classifiers are created by randomly  drawn bootstrap subsets of both examples and features. \n",
    "\n",
    "Lastly, after the base classifiers are trained, we combine them using cost-sensitive weighted voting. This method is based in based in calculating the final prediction as a weighted average of the base classifiers estimation:\n",
    "\n",
    "$$  f_{wv}(\\mathcal{S},\\mathcal{M}, \\alpha)\n",
    "  =argmax_{c \\in \\{0,1\\}} \\sum_{j=1}^T \\alpha_j \\mathbf{1}_c(M_j(\\mathcal{S})), $$\n",
    "  \n",
    "where  $\\alpha_j$ is related to the performance of each classifier $M_j$ in the  out of bag set   $\\mathcal{S}_j^{oob}=\\mathcal{S}-\\mathcal{S}_j$\n",
    "\n",
    "$$  \\alpha_j=\\frac{Savings(M_j(\\mathcal{S}_j^{oob}))}\n",
    "  {\\sum_{j_1=1}^T Savings(M_{j_1}(\\mathcal{S}_j^{oob}))}. $$\n",
    "\n",
    "This method guaranties that the base classifiers that contribute to a higher increase in savings \n",
    "have more importance in the ensemble. For more details see [20]."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>f1</th>\n",
       "      <th>pre</th>\n",
       "      <th>acc</th>\n",
       "      <th>rec</th>\n",
       "      <th>sav</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>DT</th>\n",
       "      <td>0.258659</td>\n",
       "      <td>0.245158</td>\n",
       "      <td>0.894612</td>\n",
       "      <td>0.273734</td>\n",
       "      <td>0.191818</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LR</th>\n",
       "      <td>0.021739</td>\n",
       "      <td>0.583333</td>\n",
       "      <td>0.933048</td>\n",
       "      <td>0.011076</td>\n",
       "      <td>0.003708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RF</th>\n",
       "      <td>0.238577</td>\n",
       "      <td>0.511265</td>\n",
       "      <td>0.933296</td>\n",
       "      <td>0.155591</td>\n",
       "      <td>0.127245</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DT-BMR</th>\n",
       "      <td>0.136897</td>\n",
       "      <td>0.082195</td>\n",
       "      <td>0.653371</td>\n",
       "      <td>0.409283</td>\n",
       "      <td>0.086920</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LR-BMR</th>\n",
       "      <td>0.169111</td>\n",
       "      <td>0.099707</td>\n",
       "      <td>0.632754</td>\n",
       "      <td>0.556435</td>\n",
       "      <td>0.184102</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RF-BMR</th>\n",
       "      <td>0.264445</td>\n",
       "      <td>0.166920</td>\n",
       "      <td>0.762337</td>\n",
       "      <td>0.636076</td>\n",
       "      <td>0.402713</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CSDT</th>\n",
       "      <td>0.270222</td>\n",
       "      <td>0.162937</td>\n",
       "      <td>0.712990</td>\n",
       "      <td>0.791139</td>\n",
       "      <td>0.464976</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CSRP</th>\n",
       "      <td>0.381598</td>\n",
       "      <td>0.274948</td>\n",
       "      <td>0.864288</td>\n",
       "      <td>0.623418</td>\n",
       "      <td>0.475024</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              f1       pre       acc       rec       sav\n",
       "DT      0.258659  0.245158  0.894612  0.273734  0.191818\n",
       "LR      0.021739  0.583333  0.933048  0.011076  0.003708\n",
       "RF      0.238577  0.511265  0.933296  0.155591  0.127245\n",
       "DT-BMR  0.136897  0.082195  0.653371  0.409283  0.086920\n",
       "LR-BMR  0.169111  0.099707  0.632754  0.556435  0.184102\n",
       "RF-BMR  0.264445  0.166920  0.762337  0.636076  0.402713\n",
       "CSDT    0.270222  0.162937  0.712990  0.791139  0.464976\n",
       "CSRP    0.381598  0.274948  0.864288  0.623418  0.475024"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from costcla.models import CostSensitiveRandomPatchesClassifier\n",
    "\n",
    "classifiers[\"CSRP\"] = {\"f\": CostSensitiveRandomPatchesClassifier(combination='weighted_voting')}\n",
    "# Fit\n",
    "classifiers[\"CSRP\"][\"f\"].fit(X_train, y_train, cost_mat_train)\n",
    "# Predict\n",
    "classifiers[\"CSRP\"][\"c\"] = classifiers[\"CSRP\"][\"f\"].predict(X_test)\n",
    "# Evaluate\n",
    "results.loc[\"CSRP\"] = 0\n",
    "results.loc[\"CSRP\", measures.keys()] = \\\n",
    "[measures[measure](y_test, classifiers[\"CSRP\"][\"c\"]) for measure in measures.keys()]\n",
    "results[\"sav\"].loc[\"CSRP\"] = savings_score(y_test, classifiers[\"CSRP\"][\"c\"], cost_mat_test)\n",
    "    \n",
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArUAAAE8CAYAAADe90nMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nibslAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f93c4b18358>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the results\n",
    "ind = np.arange(results.shape[0])\n",
    "figsize(10, 5)\n",
    "ax = plt.subplot(111)\n",
    "l = ax.plot(ind, results[\"f1\"], \"-o\", label='F1Score')\n",
    "b = ax.bar(ind-0.3, results['sav'], 0.6, label='Savings')\n",
    "plt.legend(loc='center left', bbox_to_anchor=(1, 0.5))\n",
    "ax.set_xlim([-0.5, ind[-1]+.5])\n",
    "ax.set_xticks(ind)\n",
    "ax.set_xticklabels(results.index)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conclusions\n",
    "\n",
    "CostCla is a easy to use Python library for example-dependent cost-sensitive \n",
    "classification problems. It includes many example-dependent cost-sensitive algorithms. Since \n",
    "it is part of the scientific Python ecosystem, it can be easily integrated with other \n",
    "machine learning libraries. Future work includes adding more cost-sensitive databases and \n",
    "algorithms, and support for Python $\\ge$ 3.4. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## References\n",
    "1. Anderson, R. (2007). The Credit Scoring Toolkit : Theory and Practice for Retail Credit Risk Management and Decision Automation. Oxford University Press.\n",
    "2. Hand, D. J., & Henley, W. E. (1997). Statistical Classification Methods in Consumer Credit Scoring: A Review. Journal of the Royal Statstical Society. Series A (Statistics in Society), 160(3), 523–541.\n",
    "3. Correa Bahnsen, A., & Gonzalez Montoya, A. (2011). Evolutionary Algorithms for Selecting the Architecture of a MLP Neural Network: A Credit Scoring Case. In IEEE 11th International Conference on Data Mining Workshops (pp. 725–732). Ieee. http://doi.org/10.1109/ICDMW.2011.80\n",
    "4. Beling, P., Covaliu, Z., & Oliver, R. M. (2005). Optimal scoring cutoff policies and efficient frontiers. Journal of the Operational Research Society, 56(9), 1016–1029. http://doi.org/10.1057/palgrave.jors.2602021\n",
    "5. Verbraken, T., Bravo, C., Weber, R., & Baesens, B. (2014). Development and application of consumer credit scoring models using profit-based classification measures. European Journal of Operational Research, 238(2), 505–513. http://doi.org/10.1016/j.ejor.2014.04.001\n",
    "6.  Alejo, R., & Garc, V. (2013). Making Accurate Credit Risk Predictions with Cost-Sensitive MLP Neural Networks. In J. Casillas, F. J. Martínez-López, R. Vicari, & F. De la Prieta (Eds.), Advances in Intelligent Systems and Computing (Vol. 220, pp. 1–8). Heidelberg: Springer International Publishing. http://doi.org/10.1007/978-3-319-00569-0\n",
    "7.  Oliver, R., & Thomas, L. (2009). Optimal score cutoffs and pricing in regulatory capital in retail credit portfolios (No. CRR-09-01). Southampton. Retrieved from http://eprints.soton.ac.uk/71321/\n",
    "8.  Correa Bahnsen, A., Aouada, D., & Ottersten, B. (2014). Example-Dependent Cost-Sensitive Logistic Regression for Credit Scoring. In 2014 13th International Conference on Machine Learning and Applications (pp. 263–269). Detroit, USA: IEEE. http://doi.org/10.1109/ICMLA.2014.48\n",
    "9.  Lawrence, D., & Solomon, A. (2012). Managing a Consumer Lending Business. Solomon Lawrence Partners. \n",
    "10. Nayak, G. N., & Turvey, C. G. (1997). Credit Risk Assessment and the Opportunity Costs of Loan Misclassification. Canadian Journal of Agricultural Economics, 45(3), 285–299. http://doi.org/10.1111/j.1744-7976.1997.tb00209.x\n",
    "11. ECB. (2014). European Central Bank. Retrieved from http://www.ecb.europa.eu/stats\n",
    "12. Jayanta K., G., Mohan, D., & Tapas, S. (2006). Bayesian Inference and Decision Theory. In An Introduction to Bayesian Analysis (Vol. 13, pp. 26–63). Springer New York. http://doi.org/10.1007/978-0-387-35433-0_2\n",
    "13. Correa Bahnsen, A., Aouada, D., & Ottersten, B. (2015). Example-Dependent Cost-Sensitive Decision Trees. Expert Systems with Applications, in press. http://doi.org/10.1016/j.eswa.2015.04.042\n",
    "14. Maimon, L. R. O. (2008). Data Mining with Decision Trees: Theory and Applications. World Scientific Publishing Company.\n",
    "15. Hastie, T., Tibshirani, R., & Friedman, J. (2009). The Elements of Statistical Learning: Data Mining, Inference, and Prediction (2nd ed.). Stanford, CA: Springer.\n",
    "16. Rokach, L., & Maimon, O. (2010). Decision Trees. In L. Rokach & O. Maimon (Eds.), Data Mining and Knowledge Discovery Handbook (2nd ed., pp. 149–174). Springer US. http://doi.org/10.1007/978-0-387-09823-4_9\n",
    "17. \\citep{Rokach2009}\n",
    "18. \\citep{Zhou2012}\n",
    "19. \\citep{Louppe2012}\n",
    "20. Correa Bahnsen, A., Aouada, D., & Ottersten, B. (2015). Ensembles of Example-Dependent Cost-Sensitive Decision Trees. http://arxiv.org/abs/1505.04637"
   ]
  }
 ],
 "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": "3.5.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}