exploratory data analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exploratory Data Analysis\n",
    "\n",
    "\n",
    "Over the past year and a half, I've worked in the industry and in academia. My introduction to exploratory data analysis began while writing my bachelor thesis in January 2015 where I worked at an electro-mobility firm in Germany trying to investigate how energy storage designs can be optimized. A 'good design' was one that affected the KPIs positively and these KPIs were often a proxy for the overall business metric of the manufacturing plant. \n",
    "\n",
    "Besides peforming simulations, questions I was being asked everyday went something like this:\n",
    "\n",
    "1. How are attributes (features) X1, X2, X3, X4 ... XN distributed?\n",
    "2. How does the response Y behave when conditioned by attribute X1? \n",
    "3. Is there an underlying pattern to the outliers? \n",
    "4. What are the interactions or correlations between attributes pairs? \n",
    "5. Which are the most contributing attributes?\n",
    "6. If certain attributes are set to their mean-value, how does the relationship between the most contributing attributes and the response look like? This is also knowing as *'adjusting the response'*. \n",
    "\n",
    "## Data is often messy due a number of reasons\n",
    "\n",
    "Fastforward to today, I've been doing energy analytics at IIT Bombay, INDIA. My job mainly requires me to work with Internet-of-Things (IoT) systems and sensor data (smart-meters).  A **sensor is a data collector**  that monitors the environment and records certain measurements. A simple example would be a luminosity sensor, it measures light intensity that hits its surface and logs these measurements in a file or a database. This data is logged into databases (MongoDB or MySQL) at frequencies of the order of milli-seconds. Anyone who has worked ground-up with sensors, will know that **sensor data is extremely messy**. This 'messiness'can creep in due to various reasons - hardware malfunctioning, data-logging errors, anamolies etc. Problems become worse when you want to fuse data from various sensors (sources) and there is a disparity between their timestamps. \n",
    "\n",
    "## Why Exploratory Data Analysis?\n",
    "\n",
    "In such cases, it is often imperative to perform exloratory analysis to understand what's going on in the underlying process that generates the data. The following excerpt from <cite> “Doing Data Science” by Rachel Schutt and Cathy O’Neil <cite> articulates this necessity very lucidly - \n",
    "\n",
    "*\"As much as Exploratory Data Analysis (EDA) is a set of tools, it is also a mindset. And that mindset is about your relationship with the data. You want to understand the data-gain intuition, understand the shape of it and try to connect your understanding of the process that generated the data to the data itself. EDA happens between you and the data and isn't about proving anything to anyone else”* \n",
    "\n",
    "**TL;DR** before building ANY machine learning model its a great habit to perform EDA and build some intuition. It can help you to -\n",
    "\n",
    "* Visualize the shape of the data.\n",
    "* How the response variable behaves when conditioned by a feature. \n",
    "* Identify correlations between features.\n",
    "* Performing sanity checks on your resultant machine learning model and troubleshoot errors.\n",
    "* Assess the scale and direction of the relationship between features and outcome variables"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Let's get our hands dirty: UCI Bank Marketing Dataset\n",
    "\n",
    "The [data](https://archive.ics.uci.edu/ml/datasets/Bank+Marketing) is related with direct marketing campaigns of a Portuguese banking institution. The marketing campaigns were based on phone calls. \n",
    "\n",
    "**The output is the conversion metric:** whether a customer signed up for the bank product. \n",
    "\n",
    "**Type of problem:** Basically, it is a **binary classification problem**.\n",
    "\n",
    "### Why this dataset? \n",
    "\n",
    "**Firstly**, because I am personally interested to see how data can be analyzed to **build data products that influence business decisions**. For example in marketing, competition and budget constraints demand that the teams invest on highly refined and directed campaigns that maximize their Return On Investment. It is especially interesting for me to see how insights pulled from data can be used to improve these campaigns. This isn't very different from  how analytics is used to optimize the design of IoT systems before scaling up but more on that later.\n",
    "\n",
    "**Secondly**, the dataset is **intuitive and requires little domain knowledge**. I've personally found that the quickest way to learn is by picking up a dataset that contains interpretable features (social and economic) so that you have build some intuition while building models and still stay connected to the overall business problem/metric. \n",
    "\n",
    "\n",
    "I will adopt a **question and answer style through this exercise**; a **high-level business question** and **answer in the form of a plot or a graph**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 237,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import pylab\n",
    "from sklearn import preprocessing\n",
    "from __future__ import division\n",
    "sns.set(style=\"darkgrid\")\n",
    "pylab.rcParams['figure.figsize'] = (8.0, 6.0)\n",
    "% matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>job</th>\n",
       "      <th>marital</th>\n",
       "      <th>education</th>\n",
       "      <th>default</th>\n",
       "      <th>balance</th>\n",
       "      <th>housing</th>\n",
       "      <th>loan</th>\n",
       "      <th>contact</th>\n",
       "      <th>day</th>\n",
       "      <th>month</th>\n",
       "      <th>duration</th>\n",
       "      <th>campaign</th>\n",
       "      <th>pdays</th>\n",
       "      <th>previous</th>\n",
       "      <th>poutcome</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>30</td>\n",
       "      <td>unemployed</td>\n",
       "      <td>married</td>\n",
       "      <td>primary</td>\n",
       "      <td>no</td>\n",
       "      <td>1787</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>cellular</td>\n",
       "      <td>19</td>\n",
       "      <td>oct</td>\n",
       "      <td>79</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>unknown</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>33</td>\n",
       "      <td>services</td>\n",
       "      <td>married</td>\n",
       "      <td>secondary</td>\n",
       "      <td>no</td>\n",
       "      <td>4789</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>cellular</td>\n",
       "      <td>11</td>\n",
       "      <td>may</td>\n",
       "      <td>220</td>\n",
       "      <td>1</td>\n",
       "      <td>339</td>\n",
       "      <td>4</td>\n",
       "      <td>failure</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>35</td>\n",
       "      <td>management</td>\n",
       "      <td>single</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>no</td>\n",
       "      <td>1350</td>\n",
       "      <td>yes</td>\n",
       "      <td>no</td>\n",
       "      <td>cellular</td>\n",
       "      <td>16</td>\n",
       "      <td>apr</td>\n",
       "      <td>185</td>\n",
       "      <td>1</td>\n",
       "      <td>330</td>\n",
       "      <td>1</td>\n",
       "      <td>failure</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>30</td>\n",
       "      <td>management</td>\n",
       "      <td>married</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>no</td>\n",
       "      <td>1476</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>unknown</td>\n",
       "      <td>3</td>\n",
       "      <td>jun</td>\n",
       "      <td>199</td>\n",
       "      <td>4</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>unknown</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>59</td>\n",
       "      <td>blue-collar</td>\n",
       "      <td>married</td>\n",
       "      <td>secondary</td>\n",
       "      <td>no</td>\n",
       "      <td>0</td>\n",
       "      <td>yes</td>\n",
       "      <td>no</td>\n",
       "      <td>unknown</td>\n",
       "      <td>5</td>\n",
       "      <td>may</td>\n",
       "      <td>226</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>unknown</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age          job  marital  education default  balance housing loan  \\\n",
       "0   30   unemployed  married    primary      no     1787      no   no   \n",
       "1   33     services  married  secondary      no     4789     yes  yes   \n",
       "2   35   management   single   tertiary      no     1350     yes   no   \n",
       "3   30   management  married   tertiary      no     1476     yes  yes   \n",
       "4   59  blue-collar  married  secondary      no        0     yes   no   \n",
       "\n",
       "    contact  day month  duration  campaign  pdays  previous poutcome   y  \n",
       "0  cellular   19   oct        79         1     -1         0  unknown  no  \n",
       "1  cellular   11   may       220         1    339         4  failure  no  \n",
       "2  cellular   16   apr       185         1    330         1  failure  no  \n",
       "3   unknown    3   jun       199         4     -1         0  unknown  no  \n",
       "4   unknown    5   may       226         1     -1         0  unknown  no  "
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv('../bank/bank.csv', sep=';')\n",
    "data_labels = pd.Series([0 if x == 'no' else 1 for x in data.ix[:,-1]])\n",
    "data.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Exploration\n",
    "\n",
    "Above, we can see there are many interpretable social features like age, contact details, marital status etc. \n",
    "\n",
    "**QUESTION: Was the campaign targeted towards a particular range of age? How did the differenly aged people convert or sign up for the product?** "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x112a3e710>"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAa0AAAKCCAYAAABiRIZJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nZYwxxjWs0zLGGOMa1mkZY4xxjf8HW5nLkyxmaroAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x112a44a10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.FacetGrid(data, row=\"contact\", col=\"y\", margin_titles=True)\n",
    "g.map(sns.distplot, \"age\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**ANSWER:** The **campaign wasn't specifically targetted** towards any specific age group. Let's have a look at the overall distribution of age-class to make this clearer. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x112268790>"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAECCAYAAAD3vwBsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAJJUKAtAkgplAUhSof4filRY2DMK1wMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10b186e10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data.age.hist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**QUESTION: How are the jobs distributed in the customer base?**  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Different categories of jobs\n",
      "\n",
      "management       969\n",
      "blue-collar      946\n",
      "technician       768\n",
      "admin.           478\n",
      "services         417\n",
      "retired          230\n",
      "self-employed    183\n",
      "entrepreneur     168\n",
      "unemployed       128\n",
      "housemaid        112\n",
      "student           84\n",
      "unknown           38\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print 'Different categories of jobs\\n'\n",
    "print data.job.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1124f4490>"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAApwAAAF8CAYAAABi52u6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11251b790>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pylab\n",
    "pylab.rcParams['figure.figsize'] = (10.0, 6.0)\n",
    "sns.countplot(y=data.job)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Answer: ** Seems like the majority of the customers where corporate blue-collar workers, management officials and technicians. Let's see how these majority classes converted.\n",
    "\n",
    "**QUESTION: How succesfully were the blue-collar workers, management workers and technicians converted to buy the product?**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Proportion of population doing majority jobs 0.593452775935\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x112528110>"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbYAAAF8CAYAAACjep6iAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAV/6AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1115c52d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pylab.rcParams['figure.figsize'] = (6.0, 6.0)\n",
    "majority_jobs = data[(data.job == 'management') | (data.job == 'technician') | (data.job == 'blue-collar')]\n",
    "print 'Proportion of population doing majority jobs', len(majority_jobs)/ (data.shape[0])\n",
    "sns.countplot(y=majority_jobs.job, hue=majority_jobs.y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**ANSWER:** As can be seen from the above graph, management workers seemed to be more interested in buying the product whereas blue-collar workers were more hesistant. Another insight is that **there seems to be a class imbalance**. These majority jobs represent 60% of the customers and there seems to be a class imbalance which can also be confirmed by making a histrogram of column y. Such class imbalances can severely affect our classification algorithm/model. More on how to handle class imbalance problems later."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conditional Probability distributions: Response conditioned by categorical features.\n",
    "\n",
    "This is an important step to determine if a particular feature has a predictive power. Earlier we deduced that the advertising campaign wasn't specifically targetted towards a specific age class. But now we want to answer:\n",
    "\n",
    "**QUESTION: What influence did age have in the decision of buying a product?**\n",
    "This can be answered by making boxplots. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x111647cd0>"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfgAAAF8CAYAAAAjCamxAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAFCQwANAQQIPAAX9fxqeUFI8Z3AVAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1124e2050>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pylab.rcParams['figure.figsize'] = (8.0, 6.0)\n",
    "sns.boxplot(y=data['y'], x=data['age'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**ANSWER:** Amongst people who bought the product (y=yes) **a larger proportion of people were above 40 years in age**. For the sake of brevity, I will also plot a kernel density estimate plot. \n",
    "\n",
    "#### KDE Plots"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfUAAAFxCAYAAABqaTHfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngiAIW4KIuiAIgiBsCf8fgI2d5zKgYRsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11164a210>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "pylab.rcParams['figure.figsize'] = (8.0, 6.0)\n",
    "sns.kdeplot(data[(data['y'] == 'no')].age, shade=True, alpha=.2, label='No', color = 'salmon')\n",
    "sns.kdeplot(data[(data['y'] == 'yes')].age, shade=True, alpha=.2, label='Yes')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There you go, much more clearer. The KDE plot has a wider tail."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Correlations!\n",
    "\n",
    "Phew finally we have a feel of how different attributes of the sample (or the population) influence their behaviour of buying the product. \n",
    "\n",
    "#### Feature-Response Correlations\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>balance</th>\n",
       "      <th>day</th>\n",
       "      <th>duration</th>\n",
       "      <th>campaign</th>\n",
       "      <th>pdays</th>\n",
       "      <th>previous</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>age</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.083820</td>\n",
       "      <td>-0.017853</td>\n",
       "      <td>-0.002367</td>\n",
       "      <td>-0.005148</td>\n",
       "      <td>-0.008894</td>\n",
       "      <td>-0.003511</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>balance</th>\n",
       "      <td>0.083820</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.008677</td>\n",
       "      <td>-0.015950</td>\n",
       "      <td>-0.009976</td>\n",
       "      <td>0.009437</td>\n",
       "      <td>0.026196</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>day</th>\n",
       "      <td>-0.017853</td>\n",
       "      <td>-0.008677</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.024629</td>\n",
       "      <td>0.160706</td>\n",
       "      <td>-0.094352</td>\n",
       "      <td>-0.059114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>duration</th>\n",
       "      <td>-0.002367</td>\n",
       "      <td>-0.015950</td>\n",
       "      <td>-0.024629</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.068382</td>\n",
       "      <td>0.010380</td>\n",
       "      <td>0.018080</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>campaign</th>\n",
       "      <td>-0.005148</td>\n",
       "      <td>-0.009976</td>\n",
       "      <td>0.160706</td>\n",
       "      <td>-0.068382</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.093137</td>\n",
       "      <td>-0.067833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pdays</th>\n",
       "      <td>-0.008894</td>\n",
       "      <td>0.009437</td>\n",
       "      <td>-0.094352</td>\n",
       "      <td>0.010380</td>\n",
       "      <td>-0.093137</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.577562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>previous</th>\n",
       "      <td>-0.003511</td>\n",
       "      <td>0.026196</td>\n",
       "      <td>-0.059114</td>\n",
       "      <td>0.018080</td>\n",
       "      <td>-0.067833</td>\n",
       "      <td>0.577562</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               age   balance       day  duration  campaign     pdays  previous\n",
       "age       1.000000  0.083820 -0.017853 -0.002367 -0.005148 -0.008894 -0.003511\n",
       "balance   0.083820  1.000000 -0.008677 -0.015950 -0.009976  0.009437  0.026196\n",
       "day      -0.017853 -0.008677  1.000000 -0.024629  0.160706 -0.094352 -0.059114\n",
       "duration -0.002367 -0.015950 -0.024629  1.000000 -0.068382  0.010380  0.018080\n",
       "campaign -0.005148 -0.009976  0.160706 -0.068382  1.000000 -0.093137 -0.067833\n",
       "pdays    -0.008894  0.009437 -0.094352  0.010380 -0.093137  1.000000  0.577562\n",
       "previous -0.003511  0.026196 -0.059114  0.018080 -0.067833  0.577562  1.000000"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#data['loan'] = data.loan.apply(lambda x: 1 if (x == 'yes') else 0)\n",
    "data.corr()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preprocessing and Dimensionality Reduction\n",
    "\n",
    "Finally, I wish to perform some dimensionality reduction. I will be applying principal component analysis for the same.\n",
    "\n",
    "*Principal Component Analysis (PCA) is a simple yet popular and useful linear transformation technique that is used in numerous applications, such as stock market predictions, the analysis of gene expression data, and many more.*\n",
    "\n",
    "But there is a slight problem here since we have categorical and numerical data. The first step for PCA is to form a covariance or the correlation matrix for the entire feature space. The strength of correlation between any pair of variables is expressed by covariance and a correlation coefficient for numerical data. \n",
    "\n",
    "But we have a non-numeric or categorical features like housing, loan, contact, education and so on. To apply PCA, I will use dummy encoding to transform categorical features to numerical features.\n",
    "\n",
    "### Dummy Encoding: Converting categorical features to numerical features.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index([u'age', u'job', u'marital', u'education', u'default', u'balance', u'housing', u'loan', u'contact', u'day', u'month', u'duration', u'campaign', u'pdays', u'previous', u'poutcome', u'y'], dtype='object')\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>job</th>\n",
       "      <th>marital</th>\n",
       "      <th>education</th>\n",
       "      <th>default</th>\n",
       "      <th>balance</th>\n",
       "      <th>housing</th>\n",
       "      <th>loan</th>\n",
       "      <th>contact</th>\n",
       "      <th>day</th>\n",
       "      <th>month</th>\n",
       "      <th>duration</th>\n",
       "      <th>campaign</th>\n",
       "      <th>pdays</th>\n",
       "      <th>previous</th>\n",
       "      <th>poutcome</th>\n",
       "      <th>y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>30</td>\n",
       "      <td>unemployed</td>\n",
       "      <td>married</td>\n",
       "      <td>primary</td>\n",
       "      <td>no</td>\n",
       "      <td>1787</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>cellular</td>\n",
       "      <td>19</td>\n",
       "      <td>oct</td>\n",
       "      <td>79</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>unknown</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>33</td>\n",
       "      <td>services</td>\n",
       "      <td>married</td>\n",
       "      <td>secondary</td>\n",
       "      <td>no</td>\n",
       "      <td>4789</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>cellular</td>\n",
       "      <td>11</td>\n",
       "      <td>may</td>\n",
       "      <td>220</td>\n",
       "      <td>1</td>\n",
       "      <td>339</td>\n",
       "      <td>4</td>\n",
       "      <td>failure</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>35</td>\n",
       "      <td>management</td>\n",
       "      <td>single</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>no</td>\n",
       "      <td>1350</td>\n",
       "      <td>yes</td>\n",
       "      <td>no</td>\n",
       "      <td>cellular</td>\n",
       "      <td>16</td>\n",
       "      <td>apr</td>\n",
       "      <td>185</td>\n",
       "      <td>1</td>\n",
       "      <td>330</td>\n",
       "      <td>1</td>\n",
       "      <td>failure</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>30</td>\n",
       "      <td>management</td>\n",
       "      <td>married</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>no</td>\n",
       "      <td>1476</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>unknown</td>\n",
       "      <td>3</td>\n",
       "      <td>jun</td>\n",
       "      <td>199</td>\n",
       "      <td>4</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>unknown</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>59</td>\n",
       "      <td>blue-collar</td>\n",
       "      <td>married</td>\n",
       "      <td>secondary</td>\n",
       "      <td>no</td>\n",
       "      <td>0</td>\n",
       "      <td>yes</td>\n",
       "      <td>no</td>\n",
       "      <td>unknown</td>\n",
       "      <td>5</td>\n",
       "      <td>may</td>\n",
       "      <td>226</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>0</td>\n",
       "      <td>unknown</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   age          job  marital  education default  balance housing loan  \\\n",
       "0   30   unemployed  married    primary      no     1787      no   no   \n",
       "1   33     services  married  secondary      no     4789     yes  yes   \n",
       "2   35   management   single   tertiary      no     1350     yes   no   \n",
       "3   30   management  married   tertiary      no     1476     yes  yes   \n",
       "4   59  blue-collar  married  secondary      no        0     yes   no   \n",
       "\n",
       "    contact  day month  duration  campaign  pdays  previous poutcome   y  \n",
       "0  cellular   19   oct        79         1     -1         0  unknown  no  \n",
       "1  cellular   11   may       220         1    339         4  failure  no  \n",
       "2  cellular   16   apr       185         1    330         1  failure  no  \n",
       "3   unknown    3   jun       199         4     -1         0  unknown  no  \n",
       "4   unknown    5   may       226         1     -1         0  unknown  no  "
      ]
     },
     "execution_count": 175,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clean_data = data.copy()\n",
    "print clean_data.columns\n",
    "clean_data.head(5)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>job</th>\n",
       "      <th>marital</th>\n",
       "      <th>education</th>\n",
       "      <th>default</th>\n",
       "      <th>housing</th>\n",
       "      <th>loan</th>\n",
       "      <th>contact</th>\n",
       "      <th>day</th>\n",
       "      <th>month</th>\n",
       "      <th>previous</th>\n",
       "      <th>poutcome</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>unemployed</td>\n",
       "      <td>married</td>\n",
       "      <td>primary</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>cellular</td>\n",
       "      <td>19</td>\n",
       "      <td>oct</td>\n",
       "      <td>0</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>services</td>\n",
       "      <td>married</td>\n",
       "      <td>secondary</td>\n",
       "      <td>no</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>cellular</td>\n",
       "      <td>11</td>\n",
       "      <td>may</td>\n",
       "      <td>4</td>\n",
       "      <td>failure</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>management</td>\n",
       "      <td>single</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>no</td>\n",
       "      <td>yes</td>\n",
       "      <td>no</td>\n",
       "      <td>cellular</td>\n",
       "      <td>16</td>\n",
       "      <td>apr</td>\n",
       "      <td>1</td>\n",
       "      <td>failure</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>management</td>\n",
       "      <td>married</td>\n",
       "      <td>tertiary</td>\n",
       "      <td>no</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>unknown</td>\n",
       "      <td>3</td>\n",
       "      <td>jun</td>\n",
       "      <td>0</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>blue-collar</td>\n",
       "      <td>married</td>\n",
       "      <td>secondary</td>\n",
       "      <td>no</td>\n",
       "      <td>yes</td>\n",
       "      <td>no</td>\n",
       "      <td>unknown</td>\n",
       "      <td>5</td>\n",
       "      <td>may</td>\n",
       "      <td>0</td>\n",
       "      <td>unknown</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           job  marital  education default housing loan   contact  day month  \\\n",
       "0   unemployed  married    primary      no      no   no  cellular   19   oct   \n",
       "1     services  married  secondary      no     yes  yes  cellular   11   may   \n",
       "2   management   single   tertiary      no     yes   no  cellular   16   apr   \n",
       "3   management  married   tertiary      no     yes  yes   unknown    3   jun   \n",
       "4  blue-collar  married  secondary      no     yes   no   unknown    5   may   \n",
       "\n",
       "   previous poutcome  \n",
       "0         0  unknown  \n",
       "1         4  failure  \n",
       "2         1  failure  \n",
       "3         0  unknown  \n",
       "4         0  unknown  "
      ]
     },
     "execution_count": 176,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_cols = [1, 2, 3, 4, 6, 7, 8, 9,10,14,15]\n",
    "cat_data = clean_data.iloc[:,cat_cols]\n",
    "cat_data.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nice! We have all the categorical features in our dataframe *cat_data*. We will now apply dummy coding to each feature of this dataframe and join/merge this transformed dataframe back to our original dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(4521, 46)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>day</th>\n",
       "      <th>previous</th>\n",
       "      <th>job_admin.</th>\n",
       "      <th>job_blue-collar</th>\n",
       "      <th>job_entrepreneur</th>\n",
       "      <th>job_housemaid</th>\n",
       "      <th>job_management</th>\n",
       "      <th>job_retired</th>\n",
       "      <th>job_self-employed</th>\n",
       "      <th>job_services</th>\n",
       "      <th>...</th>\n",
       "      <th>month_jun</th>\n",
       "      <th>month_mar</th>\n",
       "      <th>month_may</th>\n",
       "      <th>month_nov</th>\n",
       "      <th>month_oct</th>\n",
       "      <th>month_sep</th>\n",
       "      <th>poutcome_failure</th>\n",
       "      <th>poutcome_other</th>\n",
       "      <th>poutcome_success</th>\n",
       "      <th>poutcome_unknown</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>16</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 46 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   day  previous  job_admin.  job_blue-collar  job_entrepreneur  \\\n",
       "0   19         0           0                0                 0   \n",
       "1   11         4           0                0                 0   \n",
       "2   16         1           0                0                 0   \n",
       "3    3         0           0                0                 0   \n",
       "4    5         0           0                1                 0   \n",
       "\n",
       "   job_housemaid  job_management  job_retired  job_self-employed  \\\n",
       "0              0               0            0                  0   \n",
       "1              0               0            0                  0   \n",
       "2              0               1            0                  0   \n",
       "3              0               1            0                  0   \n",
       "4              0               0            0                  0   \n",
       "\n",
       "   job_services        ...         month_jun  month_mar  month_may  month_nov  \\\n",
       "0             0        ...                 0          0          0          0   \n",
       "1             1        ...                 0          0          1          0   \n",
       "2             0        ...                 0          0          0          0   \n",
       "3             0        ...                 1          0          0          0   \n",
       "4             0        ...                 0          0          1          0   \n",
       "\n",
       "   month_oct  month_sep  poutcome_failure  poutcome_other  poutcome_success  \\\n",
       "0          1          0                 0               0                 0   \n",
       "1          0          0                 1               0                 0   \n",
       "2          0          0                 1               0                 0   \n",
       "3          0          0                 0               0                 0   \n",
       "4          0          0                 0               0                 0   \n",
       "\n",
       "   poutcome_unknown  \n",
       "0                 1  \n",
       "1                 0  \n",
       "2                 0  \n",
       "3                 1  \n",
       "4                 1  \n",
       "\n",
       "[5 rows x 46 columns]"
      ]
     },
     "execution_count": 177,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dummy_cat = pd.get_dummies(cat_data)\n",
    "print dummy_cat.shape\n",
    "dummy_cat.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A total of 46 features from 10 categorical features! "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>balance</th>\n",
       "      <th>duration</th>\n",
       "      <th>campaign</th>\n",
       "      <th>pdays</th>\n",
       "      <th>y</th>\n",
       "      <th>day</th>\n",
       "      <th>previous</th>\n",
       "      <th>job_admin.</th>\n",
       "      <th>job_blue-collar</th>\n",
       "      <th>...</th>\n",
       "      <th>month_jun</th>\n",
       "      <th>month_mar</th>\n",
       "      <th>month_may</th>\n",
       "      <th>month_nov</th>\n",
       "      <th>month_oct</th>\n",
       "      <th>month_sep</th>\n",
       "      <th>poutcome_failure</th>\n",
       "      <th>poutcome_other</th>\n",
       "      <th>poutcome_success</th>\n",
       "      <th>poutcome_unknown</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>30</td>\n",
       "      <td>1787</td>\n",
       "      <td>79</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>no</td>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>33</td>\n",
       "      <td>4789</td>\n",
       "      <td>220</td>\n",
       "      <td>1</td>\n",
       "      <td>339</td>\n",
       "      <td>no</td>\n",
       "      <td>11</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>35</td>\n",
       "      <td>1350</td>\n",
       "      <td>185</td>\n",
       "      <td>1</td>\n",
       "      <td>330</td>\n",
       "      <td>no</td>\n",
       "      <td>16</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>30</td>\n",
       "      <td>1476</td>\n",
       "      <td>199</td>\n",
       "      <td>4</td>\n",
       "      <td>-1</td>\n",
       "      <td>no</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>59</td>\n",
       "      <td>0</td>\n",
       "      <td>226</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>no</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 52 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   age  balance  duration  campaign  pdays   y  day  previous  job_admin.  \\\n",
       "0   30     1787        79         1     -1  no   19         0           0   \n",
       "1   33     4789       220         1    339  no   11         4           0   \n",
       "2   35     1350       185         1    330  no   16         1           0   \n",
       "3   30     1476       199         4     -1  no    3         0           0   \n",
       "4   59        0       226         1     -1  no    5         0           0   \n",
       "\n",
       "   job_blue-collar        ...         month_jun  month_mar  month_may  \\\n",
       "0                0        ...                 0          0          0   \n",
       "1                0        ...                 0          0          1   \n",
       "2                0        ...                 0          0          0   \n",
       "3                0        ...                 1          0          0   \n",
       "4                1        ...                 0          0          1   \n",
       "\n",
       "   month_nov  month_oct  month_sep  poutcome_failure  poutcome_other  \\\n",
       "0          0          1          0                 0               0   \n",
       "1          0          0          0                 1               0   \n",
       "2          0          0          0                 1               0   \n",
       "3          0          0          0                 0               0   \n",
       "4          0          0          0                 0               0   \n",
       "\n",
       "   poutcome_success  poutcome_unknown  \n",
       "0                 0                 1  \n",
       "1                 0                 0  \n",
       "2                 0                 0  \n",
       "3                 0                 1  \n",
       "4                 0                 1  \n",
       "\n",
       "[5 rows x 52 columns]"
      ]
     },
     "execution_count": 178,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clean_data.drop(clean_data.columns[cat_cols], axis=1, inplace=True)\n",
    "clean_data = clean_data.merge(dummy_cat, how='inner',left_index=True, right_index=True, copy=False)\n",
    "clean_data.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looks good! Now time to standardize the numerical features. \n",
    "\n",
    "### Standardization of the remaining numerical features\n",
    "\n",
    "Standaridization is important to make sure we don't have overwhelming contributions of any variable due to scale."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>age</th>\n",
       "      <th>balance</th>\n",
       "      <th>duration</th>\n",
       "      <th>campaign</th>\n",
       "      <th>pdays</th>\n",
       "      <th>day</th>\n",
       "      <th>previous</th>\n",
       "      <th>job_admin.</th>\n",
       "      <th>job_blue-collar</th>\n",
       "      <th>job_entrepreneur</th>\n",
       "      <th>...</th>\n",
       "      <th>month_jun</th>\n",
       "      <th>month_mar</th>\n",
       "      <th>month_may</th>\n",
       "      <th>month_nov</th>\n",
       "      <th>month_oct</th>\n",
       "      <th>month_sep</th>\n",
       "      <th>poutcome_failure</th>\n",
       "      <th>poutcome_other</th>\n",
       "      <th>poutcome_success</th>\n",
       "      <th>poutcome_unknown</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.056270</td>\n",
       "      <td>1787</td>\n",
       "      <td>79</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.772583</td>\n",
       "      <td>4789</td>\n",
       "      <td>220</td>\n",
       "      <td>1</td>\n",
       "      <td>339</td>\n",
       "      <td>11</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.583458</td>\n",
       "      <td>1350</td>\n",
       "      <td>185</td>\n",
       "      <td>1</td>\n",
       "      <td>330</td>\n",
       "      <td>16</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-1.056270</td>\n",
       "      <td>1476</td>\n",
       "      <td>199</td>\n",
       "      <td>4</td>\n",
       "      <td>-1</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.686036</td>\n",
       "      <td>0</td>\n",
       "      <td>226</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 51 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        age  balance  duration  campaign  pdays  day  previous  job_admin.  \\\n",
       "0 -1.056270     1787        79         1     -1   19         0           0   \n",
       "1 -0.772583     4789       220         1    339   11         4           0   \n",
       "2 -0.583458     1350       185         1    330   16         1           0   \n",
       "3 -1.056270     1476       199         4     -1    3         0           0   \n",
       "4  1.686036        0       226         1     -1    5         0           0   \n",
       "\n",
       "   job_blue-collar  job_entrepreneur        ...         month_jun  month_mar  \\\n",
       "0                0                 0        ...                 0          0   \n",
       "1                0                 0        ...                 0          0   \n",
       "2                0                 0        ...                 0          0   \n",
       "3                0                 0        ...                 1          0   \n",
       "4                1                 0        ...                 0          0   \n",
       "\n",
       "   month_may  month_nov  month_oct  month_sep  poutcome_failure  \\\n",
       "0          0          0          1          0                 0   \n",
       "1          1          0          0          0                 1   \n",
       "2          0          0          0          0                 1   \n",
       "3          0          0          0          0                 0   \n",
       "4          1          0          0          0                 0   \n",
       "\n",
       "   poutcome_other  poutcome_success  poutcome_unknown  \n",
       "0               0                 0                 1  \n",
       "1               0                 0                 0  \n",
       "2               0                 0                 0  \n",
       "3               0                 0                 1  \n",
       "4               0                 0                 1  \n",
       "\n",
       "[5 rows x 51 columns]"
      ]
     },
     "execution_count": 180,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "numeric_cols = [0,11,12,13,16,17,18,19]\n",
    "data_to_scale = clean_data.iloc[:, numeric_cols].astype(np.float)  # change int to float\n",
    "clean_data.drop('y', axis=1, inplace=True)\n",
    "scaler = preprocessing.StandardScaler().fit(data_to_scale)\n",
    "clean_data.iloc[:, numeric_cols] = scaler.transform(data_to_scale)\n",
    "clean_data.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Principal Component Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn.cross_validation import train_test_split\n",
    "from sklearn.decomposition import PCA\n",
    "\n",
    "labels = pd.Series([0 if x == 'no' else 1 for x in data.ix[:,-1]])\n",
    "#clean_data.drop('y', axis=1, inplace=True)\n",
    "X = clean_data.values\n",
    "y = labels.values\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 260,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[  3.64585506e+02,  -1.84648858e+02,  -4.00210576e+01,\n",
       "          2.49976465e+00,  -2.23634221e+00],\n",
       "       [  3.36649391e+03,  -3.78625657e+01,   2.98414385e+02,\n",
       "         -2.59883566e+00,  -7.64922347e-01],\n",
       "       [ -7.24569264e+01,  -7.76755455e+01,   2.90619253e+02,\n",
       "          2.22424105e+00,  -1.12012836e+00],\n",
       "       ..., \n",
       "       [ -1.12751290e+03,  -1.14726603e+02,  -3.99147812e+01,\n",
       "          3.18269207e+00,   7.78740929e+00],\n",
       "       [ -2.85416255e+02,  -1.34531730e+02,   1.72053509e+02,\n",
       "         -8.56735562e+00,   2.15422724e+00],\n",
       "       [ -2.86703814e+02,   8.16498743e+01,   2.09087219e+02,\n",
       "         -1.12152498e+01,   5.80300940e-01]])"
      ]
     },
     "execution_count": 260,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca = PCA(n_components=5)\n",
    "pca.fit_transform(clean_data.values)\n",
    "#pca_df = pd.DataFrame(pca.components_, columns=clean_data.columns,index = ['PC-1','PC-2'])\n",
    "#np.sort(np.abs(pca_df.max(axis=0)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}