SMUMSE_ProgWorkshop_Lesson6.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Final Lesson - Introduction to Machine Learning\n",
    "\n",
    "This lesson is dedicated solely to Machine Learning. What is Machine Learning? According to Wikipedia, [Machine Learning](https://en.wikipedia.org/wiki/Machine_learning) is the scientific study of algorithms and statistical models that computer systems use in order to perform a specific task effectively without using explicit instructions, relying on patterns and inference instead.\n",
    "\n",
    "### Types of Machine Learning problems\n",
    "Typically, there are 2 kinds of machine learning problems: Supervised and Unsupervised (there is actually a third field called Reinforcement Learning, but that's not the subject of focus):\n",
    "\n",
    "From Wikipedia:\n",
    "\n",
    "    Supervised learning is the machine learning task of learning a function that maps an input to an output based \n",
    "    on example input-output pairs\n",
    "    \n",
    "    Unsupervised learning is a type of self-organized Hebbian learning that helps find previously unknown patterns \n",
    "    in data set without pre-existing labels.\n",
    "\n",
    "We will only be going through Supervised Learning problems in this lesson. That is, we are given something to predict, and we have some features that we can use to generate predictions.\n",
    "\n",
    "In the field of supervised learning problems, there are 3 subfields:\n",
    "\n",
    "1. Classification problems - In classification problems, we are given a category to predict. Examples of such questions are: is this picture a cat, a dog, or a rabbit? Is the person going to default on his loan? Given an image, predict what number it is.\n",
    "\n",
    "2. Regression problems - In regression problems, our algorithm outputs a real-valued number. Examples of such problems are: predict the price of this house, or predict GDP growth levels next year.\n",
    "\n",
    "### On Machine Learning vs Econometrics\n",
    "A better question is, how does Machine Learning differ from traditional Econometrics? According to econometrician [Francis X. Diebold](https://fxdiebold.blogspot.com/2016/10/machine-learning-vs-econometrics-i.html), Machine learning is almost always centered on prediction; think \"ŷ \", while Econometrics (E) is often, but not always, centered on prediction. Instead it's also often interested on estimation and associated inference; think \"β̂ \". Nowadays, however, the difference between the two fields seem more minute, and ML is attempting to provide answers to questions previously reserved for Econometrics.\n",
    "\n",
    "#### Disclaimer\n",
    "By no means is this a full treatment of Machine Learning; this is a gentle introduction into the wide field. For a deeper understanding of Machine Learning, online training courses hosted by Coursera and Codeacademy are recommended. In fact, this iPython notebook is derived of 3 primary components:\n",
    "\n",
    "1. Data Manipulation, Processing and Cleaning\n",
    "2. Feature Generation\n",
    "3. Machine Learning\n",
    "\n",
    "For the machine learning aspect, we will be learning two different algorithms on how to classify categorical problems. Finally, we will give some intuition into how Neural Networks work, as well as other training problems for you to work on in your free time to gain a better understanding of the other ML algorithms."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Case-Study approach\n",
    "\n",
    "In this lesson, we will not be going through the different kinds of algorithms used by Data Scientists; there are way too many algorithms. Instead, the focus will be on the important aspects before we apply the algorithms: \n",
    "\n",
    "1. Data Manipulation, Processing and Cleaning \n",
    "2. Imputing Missing Data\n",
    "3. Feature Generation\n",
    "\n",
    "The first and second section relies on much that we have learnt in lesson 3 on data manipulation, while the last section relies much on the 4th lesson. Without further ado, let's apply our understanding on the Titanic dataset.\n",
    "\n",
    "##### Training vs Testing\n",
    "In Machine Learning problems, we are typically given a training set (for us to train our algorithms), which contains the variable we are interested in, and a test set (to evaluate our predictions). Typically, the problem will include a metric for basis of comparison. \n",
    "\n",
    "##### Evaluation Metric\n",
    "For classification problems, it could be accuracy (how many test observations did you correctly predict). For regression problems, it could be RMSE (root mean-squared error).\n",
    "\n",
    "For the [Titanic problem listed on Kaggle](https://www.kaggle.com/c/titanic/overview/evaluation), they will be using \"Accuracy\" as the evaluation metric."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Preliminary Data Exploration\n",
    "\n",
    "In this section, we import key libraries and read the training and test data into Python. We look at summary statistics to get a better understanding of what we are currently dealing with."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def read_and_combine_data(train_filepath, test_filepath):\n",
    "    train = pd.read_csv(train_filepath, index_col=0)\n",
    "    test = pd.read_csv(test_filepath, index_col=0)\n",
    "    \n",
    "    combine = pd.concat([train, test], sort=True)\n",
    "    return combine\n",
    "\n",
    "train_filepath = 'data/titanic/train.csv'\n",
    "test_filepath = 'data/titanic/test.csv'\n",
    "\n",
    "train = pd.read_csv(train_filepath, index_col=0)\n",
    "test = pd.read_csv(test_filepath, index_col=0)\n",
    "combine = read_and_combine_data(train_filepath, test_filepath)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Name</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Ticket</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>22.0</td>\n",
       "      <td>N</td>\n",
       "      <td>S</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>male</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>A/5 21171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>38.0</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>PC 17599</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>26.0</td>\n",
       "      <td>N</td>\n",
       "      <td>S</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>female</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>35.0</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>113803</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>35.0</td>\n",
       "      <td>N</td>\n",
       "      <td>S</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>male</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>373450</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Age Cabin Embarked     Fare  \\\n",
       "PassengerId                                 \n",
       "1            22.0     N        S   7.2500   \n",
       "2            38.0   C85        C  71.2833   \n",
       "3            26.0     N        S   7.9250   \n",
       "4            35.0  C123        S  53.1000   \n",
       "5            35.0     N        S   8.0500   \n",
       "\n",
       "                                                          Name  Parch  Pclass  \\\n",
       "PassengerId                                                                     \n",
       "1                                      Braund, Mr. Owen Harris      0       3   \n",
       "2            Cumings, Mrs. John Bradley (Florence Briggs Th...      0       1   \n",
       "3                                       Heikkinen, Miss. Laina      0       3   \n",
       "4                 Futrelle, Mrs. Jacques Heath (Lily May Peel)      0       1   \n",
       "5                                     Allen, Mr. William Henry      0       3   \n",
       "\n",
       "                Sex  SibSp  Survived            Ticket  \n",
       "PassengerId                                             \n",
       "1              male      1       0.0         A/5 21171  \n",
       "2            female      1       1.0          PC 17599  \n",
       "3            female      0       1.0  STON/O2. 3101282  \n",
       "4            female      1       1.0            113803  \n",
       "5              male      0       0.0            373450  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combine.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Name</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Ticket</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1046.000000</td>\n",
       "      <td>295</td>\n",
       "      <td>1307</td>\n",
       "      <td>1308.000000</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309.000000</td>\n",
       "      <td>1309.000000</td>\n",
       "      <td>1309</td>\n",
       "      <td>1309.000000</td>\n",
       "      <td>891.000000</td>\n",
       "      <td>1309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>NaN</td>\n",
       "      <td>186</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1307</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>929</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>NaN</td>\n",
       "      <td>C23 C25 C27</td>\n",
       "      <td>S</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Connolly, Miss. Kate</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>male</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>CA. 2343</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>NaN</td>\n",
       "      <td>6</td>\n",
       "      <td>914</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>843</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>29.881138</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>33.295479</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.385027</td>\n",
       "      <td>2.294882</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.498854</td>\n",
       "      <td>0.383838</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>14.413493</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>51.758668</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.865560</td>\n",
       "      <td>0.837836</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.041658</td>\n",
       "      <td>0.486592</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.170000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>21.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7.895800</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>28.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>14.454200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>39.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>31.275000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>80.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>512.329200</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Age        Cabin Embarked         Fare                  Name  \\\n",
       "count   1046.000000          295     1307  1308.000000                  1309   \n",
       "unique          NaN          186        3          NaN                  1307   \n",
       "top             NaN  C23 C25 C27        S          NaN  Connolly, Miss. Kate   \n",
       "freq            NaN            6      914          NaN                     2   \n",
       "mean      29.881138          NaN      NaN    33.295479                   NaN   \n",
       "std       14.413493          NaN      NaN    51.758668                   NaN   \n",
       "min        0.170000          NaN      NaN     0.000000                   NaN   \n",
       "25%       21.000000          NaN      NaN     7.895800                   NaN   \n",
       "50%       28.000000          NaN      NaN    14.454200                   NaN   \n",
       "75%       39.000000          NaN      NaN    31.275000                   NaN   \n",
       "max       80.000000          NaN      NaN   512.329200                   NaN   \n",
       "\n",
       "              Parch       Pclass   Sex        SibSp    Survived    Ticket  \n",
       "count   1309.000000  1309.000000  1309  1309.000000  891.000000      1309  \n",
       "unique          NaN          NaN     2          NaN         NaN       929  \n",
       "top             NaN          NaN  male          NaN         NaN  CA. 2343  \n",
       "freq            NaN          NaN   843          NaN         NaN        11  \n",
       "mean       0.385027     2.294882   NaN     0.498854    0.383838       NaN  \n",
       "std        0.865560     0.837836   NaN     1.041658    0.486592       NaN  \n",
       "min        0.000000     1.000000   NaN     0.000000    0.000000       NaN  \n",
       "25%        0.000000     2.000000   NaN     0.000000    0.000000       NaN  \n",
       "50%        0.000000     3.000000   NaN     0.000000    0.000000       NaN  \n",
       "75%        0.000000     3.000000   NaN     1.000000    1.000000       NaN  \n",
       "max        9.000000     3.000000   NaN     8.000000    1.000000       NaN  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combine.describe(include='all')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Imputing Missing Data\n",
    "\n",
    "In our case, we are tasked to predict whether a passenger survived based on the features in the model. However, some of these features are not as useful as others, and we can generate new features using a combination of different features. Before we do that however, we check for missing data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Age          263\n",
       "Cabin       1014\n",
       "Embarked       2\n",
       "Fare           1\n",
       "Name           0\n",
       "Parch          0\n",
       "Pclass         0\n",
       "Sex            0\n",
       "SibSp          0\n",
       "Survived     418\n",
       "Ticket         0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum(pd.isnull(combine))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We note that there are quite a significant number of observations with missing data. To circumvent this, we can create a new column for the \"Cabin\" variable, to indicate that the data are missing. For the \"Embarked\" and \"Fare\" feature, we can impute the most popular embarkation point and the average fare."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "S    914\n",
       "C    270\n",
       "Q    123\n",
       "Name: Embarked, dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combine.Embarked.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "33.2954792813456"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combine.Fare.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def impute_missing_cabins():\n",
    "    return combine.Cabin.fillna(\"N\", inplace=True)\n",
    "\n",
    "def impute_missing_embarkation_point():\n",
    "    return combine.Embarked.fillna(\"S\", inplace=True)\n",
    "\n",
    "def impute_missing_fare():\n",
    "    return combine.Fare.fillna(combine.Fare.mean(), inplace=True)\n",
    "\n",
    "impute_missing_cabins()\n",
    "impute_missing_embarkation_point()\n",
    "impute_missing_fare()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Age         263\n",
       "Cabin         0\n",
       "Embarked      0\n",
       "Fare          0\n",
       "Name          0\n",
       "Parch         0\n",
       "Pclass        0\n",
       "Sex           0\n",
       "SibSp         0\n",
       "Survived    418\n",
       "Ticket        0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum(pd.isnull(combine))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We are left with the age variable. There are a few ways to resolve this issue:\n",
    "\n",
    "1. Drop the age variable.\n",
    "2. Impute the missing ages with the average age of the dataset\n",
    "3. Use conditional information to determine the missing age\n",
    "\n",
    "Of the 3, the third way is the hardest - we will do the third way. \n",
    "\n",
    "We first split the Name column into First and Last Name, bearing in mind that the designation of the passenger may contain some important information on the age of the passenger."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "def get_desg_first_name(name):\n",
    "    return name.split(\", \")[1]\n",
    "\n",
    "def get_first_name():\n",
    "    first_name = combine.Name.apply(lambda x: get_desg_first_name(x))\n",
    "    combine['FirstName'] = first_name.apply(lambda x: x.split(\". \")[1])\n",
    "    combine['Designation'] = first_name.apply(lambda x: x.split(\". \")[0])\n",
    "    return combine\n",
    "    \n",
    "def get_last_name():\n",
    "    combine['LastName'] = combine.Name.apply(lambda x: x.split(\", \")[0])\n",
    "    return combine\n",
    "\n",
    "get_last_name()\n",
    "get_first_name()\n",
    "print('')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Name</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Survived</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>LastName</th>\n",
       "      <th>FirstName</th>\n",
       "      <th>Designation</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>22.0</td>\n",
       "      <td>N</td>\n",
       "      <td>S</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>male</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>Braund</td>\n",
       "      <td>Owen Harris</td>\n",
       "      <td>Mr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>38.0</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>Cumings</td>\n",
       "      <td>John Bradley (Florence Briggs Thayer)</td>\n",
       "      <td>Mrs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>26.0</td>\n",
       "      <td>N</td>\n",
       "      <td>S</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>female</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>Heikkinen</td>\n",
       "      <td>Laina</td>\n",
       "      <td>Miss</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>35.0</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>113803</td>\n",
       "      <td>Futrelle</td>\n",
       "      <td>Jacques Heath (Lily May Peel)</td>\n",
       "      <td>Mrs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>35.0</td>\n",
       "      <td>N</td>\n",
       "      <td>S</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>male</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>373450</td>\n",
       "      <td>Allen</td>\n",
       "      <td>William Henry</td>\n",
       "      <td>Mr</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Age Cabin Embarked     Fare  \\\n",
       "PassengerId                                 \n",
       "1            22.0     N        S   7.2500   \n",
       "2            38.0   C85        C  71.2833   \n",
       "3            26.0     N        S   7.9250   \n",
       "4            35.0  C123        S  53.1000   \n",
       "5            35.0     N        S   8.0500   \n",
       "\n",
       "                                                          Name  Parch  Pclass  \\\n",
       "PassengerId                                                                     \n",
       "1                                      Braund, Mr. Owen Harris      0       3   \n",
       "2            Cumings, Mrs. John Bradley (Florence Briggs Th...      0       1   \n",
       "3                                       Heikkinen, Miss. Laina      0       3   \n",
       "4                 Futrelle, Mrs. Jacques Heath (Lily May Peel)      0       1   \n",
       "5                                     Allen, Mr. William Henry      0       3   \n",
       "\n",
       "                Sex  SibSp  Survived            Ticket   LastName  \\\n",
       "PassengerId                                                         \n",
       "1              male      1       0.0         A/5 21171     Braund   \n",
       "2            female      1       1.0          PC 17599    Cumings   \n",
       "3            female      0       1.0  STON/O2. 3101282  Heikkinen   \n",
       "4            female      1       1.0            113803   Futrelle   \n",
       "5              male      0       0.0            373450      Allen   \n",
       "\n",
       "                                         FirstName Designation  \n",
       "PassengerId                                                     \n",
       "1                                      Owen Harris          Mr  \n",
       "2            John Bradley (Florence Briggs Thayer)         Mrs  \n",
       "3                                            Laina        Miss  \n",
       "4                    Jacques Heath (Lily May Peel)         Mrs  \n",
       "5                                    William Henry          Mr  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combine.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that the column, \"Designation\" can be used to impute missing data for the \"Age\" variable. Suppose we have a missing data with the designation, \"Miss\". We can look at the mean value for all observations with the same designation and impute that as the missing value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Designation\n",
       "Capt            70.000000\n",
       "Col             54.000000\n",
       "Don             40.000000\n",
       "Dona            39.000000\n",
       "Dr              43.571429\n",
       "Jonkheer        38.000000\n",
       "Lady            48.000000\n",
       "Major           48.500000\n",
       "Master           5.482642\n",
       "Miss            21.774238\n",
       "Mlle            24.000000\n",
       "Mme             24.000000\n",
       "Mr              32.252151\n",
       "Mrs             36.994118\n",
       "Ms              28.000000\n",
       "Rev             41.250000\n",
       "Sir             49.000000\n",
       "the Countess    33.000000\n",
       "Name: Age, dtype: float64"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combine.groupby(\"Designation\")['Age'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Designation\n",
       "Capt              1\n",
       "Col               4\n",
       "Don               1\n",
       "Dona              1\n",
       "Dr                8\n",
       "Jonkheer          1\n",
       "Lady              1\n",
       "Major             2\n",
       "Master           61\n",
       "Miss            260\n",
       "Mlle              2\n",
       "Mme               1\n",
       "Mr              757\n",
       "Mrs             197\n",
       "Ms                2\n",
       "Rev               8\n",
       "Sir               1\n",
       "the Countess      1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combine.groupby(\"Designation\").size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PassengerId\n",
       "1         Mr\n",
       "2        Mrs\n",
       "3       Miss\n",
       "4        Mrs\n",
       "5         Mr\n",
       "6         Mr\n",
       "7         Mr\n",
       "8     Master\n",
       "9        Mrs\n",
       "10       Mrs\n",
       "11      Miss\n",
       "12      Miss\n",
       "13        Mr\n",
       "14        Mr\n",
       "15      Miss\n",
       "16       Mrs\n",
       "17    Master\n",
       "18        Mr\n",
       "19       Mrs\n",
       "20       Mrs\n",
       "Name: Designation, dtype: object"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combine.Designation[:20]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "def impute_missing_age():\n",
    "    '''\n",
    "    Generate a pd.Series using information on Designation to impute missing age data\n",
    "    '''\n",
    "    age_list = []\n",
    "    for age, desg in zip(combine.Age, combine.Designation):\n",
    "        if pd.isnull(age):\n",
    "            age_list.append(combine.groupby(\"Designation\")['Age'].mean()[desg])\n",
    "        else: age_list.append(age)\n",
    "    return age_list\n",
    "\n",
    "combine['NewAge'] = impute_missing_age()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We check for missing data once again. If everything turns out to be fine, we can begin to explore the relationships inherent in the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Age            263\n",
       "Cabin            0\n",
       "Embarked         0\n",
       "Fare             0\n",
       "Name             0\n",
       "Parch            0\n",
       "Pclass           0\n",
       "Sex              0\n",
       "SibSp            0\n",
       "Survived       418\n",
       "Ticket           0\n",
       "LastName         0\n",
       "FirstName        0\n",
       "Designation      0\n",
       "NewAge           0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum(pd.isnull(combine))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data Exploration and Plotting\n",
    "Everything seems good. We can proceed with data exploration using plots. We begin with some hypotheses that we are interested in, and see what that leads to. Here are some possible hypotheses one may be interested in testing:\n",
    "\n",
    "1. Age is correlated with Survival Rate\n",
    "2. Fare is correlated with Survival Rate\n",
    "3. Designation is correlated with Survival Rate\n",
    "\n",
    "From there, we can generate additional hypotheses: can an interaction term between Gender x Age lead to even better results? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJEAAALJCAYAAAAEQ7ISAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1091fe080>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Hypothesis 1\n",
    "plt.figure(figsize=(20, 12))\n",
    "sns.distplot(combine[combine.Survived == 0].NewAge, bins=range(0, 81, 1), color = 'red', label='Perished')\n",
    "sns.distplot(combine[combine.Survived == 1].NewAge, bins=range(0, 81, 1), color = 'blue', label='Survived')\n",
    "plt.legend(loc='upper right')\n",
    "plt.title(\"Age Distribution\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIsAAALJCAYAAAA0z6MzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11762ea58>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Hypothesis 2\n",
    "plt.figure(figsize=(20, 12))\n",
    "sns.distplot(combine[combine.Survived == 0].Fare, bins=range(0, 81, 1), color = 'red', label='Perished')\n",
    "sns.distplot(combine[combine.Survived == 1].Fare, bins=range(0, 81, 1), color = 'blue', label='Survived')\n",
    "plt.legend(loc='upper right')\n",
    "plt.title(\"Age Distribution\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the graphs above, it appears that passengers who are young tend to survive the Titanic crash. In addition, passengers who paid more for their fare tended to have a higher rate of survival. However, something stands out. There appears to be anomaly in the dataset - some passengers paid way more than \\$100 for their fare.\n",
    "\n",
    "We **may** need to conduct a secondary analysis to ensure that this is indeed an anomaly.\n",
    "\n",
    "---\n",
    "\n",
    "Let's test the third hypothesis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJUAAAK9CAYAAACdG12/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nmdXNjUpmZmZmZmZmZla3/wcDkrPXxCaEwAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x118202128>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Hypothesis 3\n",
    "plt.figure(figsize=(20, 12))\n",
    "sns.countplot(\"Designation\", hue='Survived', data=combine)\n",
    "plt.legend(loc='upper right')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The above plot seems to imply the following:\n",
    "\n",
    "1. Male passengers have a relatively low survival rate, compared to females\n",
    "2. Children tend to have a relatively high survival rate (their Designation is Master) \n",
    "3. There is not enough data about the other Designations to draw a decisive conclusion\n",
    "\n",
    "---\n",
    "\n",
    "Other hypotheses that we can carry out are as follows:\n",
    "\n",
    "1. Could the feature Ticket be correlated with the Survival Variable?\n",
    "2. Are passengers with Siblings/Spouses and Parents/Children more likely to survive than passengers travelling alone?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "929"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(np.unique(combine.Ticket))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1309"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(combine.Ticket)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We note that there are 929 unique value for tickets, but there are 1309 entries. Some passengers might be sharing tickets! To determine how many people are sharing tickets, we can use the DataFrame method, `groupby`. This can be done using the following function:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ticket\n",
       "110152               3\n",
       "110413               3\n",
       "110465               2\n",
       "110469               1\n",
       "110489               1\n",
       "110564               1\n",
       "110813               2\n",
       "111163               1\n",
       "111240               1\n",
       "111320               1\n",
       "111361               2\n",
       "111369               1\n",
       "111426               1\n",
       "111427               1\n",
       "111428               1\n",
       "112050               1\n",
       "112051               1\n",
       "112052               1\n",
       "112053               1\n",
       "112058               2\n",
       "112059               1\n",
       "112277               1\n",
       "112377               1\n",
       "112378               2\n",
       "112379               1\n",
       "112901               1\n",
       "113028               1\n",
       "113038               1\n",
       "113043               1\n",
       "113044               1\n",
       "                    ..\n",
       "STON/O 2. 3101273    1\n",
       "STON/O 2. 3101274    1\n",
       "STON/O 2. 3101275    1\n",
       "STON/O 2. 3101280    1\n",
       "STON/O 2. 3101285    1\n",
       "STON/O 2. 3101286    1\n",
       "STON/O 2. 3101288    1\n",
       "STON/O 2. 3101289    1\n",
       "STON/O 2. 3101291    1\n",
       "STON/O 2. 3101292    1\n",
       "STON/O 2. 3101293    1\n",
       "STON/O 2. 3101294    1\n",
       "STON/O2. 3101270     1\n",
       "STON/O2. 3101271     1\n",
       "STON/O2. 3101279     2\n",
       "STON/O2. 3101282     1\n",
       "STON/O2. 3101283     1\n",
       "STON/O2. 3101290     1\n",
       "STON/OQ. 369943      1\n",
       "SW/PP 751            1\n",
       "W./C. 14258          1\n",
       "W./C. 14260          1\n",
       "W./C. 14263          1\n",
       "W./C. 14266          1\n",
       "W./C. 6607           4\n",
       "W./C. 6608           5\n",
       "W./C. 6609           1\n",
       "W.E.P. 5734          2\n",
       "W/C 14208            1\n",
       "WE/P 5735            2\n",
       "Length: 929, dtype: int64"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combine.groupby('Ticket').size()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We note that there are 3 people sharing the same ticket, \"110152\". The following function returns the number of passengers holding the same ticket owned by the passenger (it's quite a difficult function to understand):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PassengerId\n",
       "1    1\n",
       "2    2\n",
       "3    1\n",
       "4    2\n",
       "5    1\n",
       "Name: Name, dtype: int64"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combine.groupby('Ticket')['Name'].transform(\"count\").head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABCYAAAJQCAYAAABSAGXZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBAAAADBGmAAAAADGCBMAAADAmP8PN8pMBG8W8NQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1091edba8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "combine['SharedTickets'] = combine.groupby('Ticket')['Name'].transform(\"count\")\n",
    "\n",
    "plt.figure(figsize=(18, 10))\n",
    "sns.countplot(combine.SharedTickets)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Are passengers who share tickets more or less likely to perish? To find out, we can use the `sns.distplot` function to find out."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIQAAAJcCAYAAACIb39OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10a735e80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20, 10))\n",
    "\n",
    "sns.distplot(combine[(combine.Survived == 0)].SharedTickets.dropna(),\n",
    "             bins = range(1, 12, 1), kde = False, norm_hist = True, color = 'red', label='Perished')\n",
    "\n",
    "sns.distplot(combine[(combine.Survived == 1)].SharedTickets.dropna(),\n",
    "             bins = range(1, 12, 1), kde = False, norm_hist = True, color = 'blue', label='Survived')\n",
    "\n",
    "plt.legend(loc='upper right')\n",
    "plt.title('Distribution of Shared Tickets')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It appears that passengers who shared tickets with 1, 2 or 3 other friends or relative had a higher chance of survival. To capture this, we can create a new feature called \"GoodSharedTicket\". In addition, we can also create a new column called, \"Alone\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "combine['GoodSharedTickets'] = ((combine.SharedTickets >= 2) & (combine.SharedTickets <= 4))\n",
    "combine['Alone'] = combine.SharedTickets == 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Name</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Survived</th>\n",
       "      <th>...</th>\n",
       "      <th>LastName</th>\n",
       "      <th>FirstName</th>\n",
       "      <th>Designation</th>\n",
       "      <th>NewAge</th>\n",
       "      <th>SharedTickets</th>\n",
       "      <th>NormalisedFare</th>\n",
       "      <th>CabinType</th>\n",
       "      <th>NoCabin</th>\n",
       "      <th>GoodSharedTickets</th>\n",
       "      <th>Alone</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>22.0</td>\n",
       "      <td>N</td>\n",
       "      <td>S</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>male</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Braund</td>\n",
       "      <td>Owen Harris</td>\n",
       "      <td>Mr</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>7.25000</td>\n",
       "      <td>N</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>38.0</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Cumings</td>\n",
       "      <td>John Bradley (Florence Briggs Thayer)</td>\n",
       "      <td>Mrs</td>\n",
       "      <td>38.0</td>\n",
       "      <td>2</td>\n",
       "      <td>35.64165</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>26.0</td>\n",
       "      <td>N</td>\n",
       "      <td>S</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>female</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Heikkinen</td>\n",
       "      <td>Laina</td>\n",
       "      <td>Miss</td>\n",
       "      <td>26.0</td>\n",
       "      <td>1</td>\n",
       "      <td>7.92500</td>\n",
       "      <td>N</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>35.0</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Futrelle</td>\n",
       "      <td>Jacques Heath (Lily May Peel)</td>\n",
       "      <td>Mrs</td>\n",
       "      <td>35.0</td>\n",
       "      <td>2</td>\n",
       "      <td>26.55000</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>35.0</td>\n",
       "      <td>N</td>\n",
       "      <td>S</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>male</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Allen</td>\n",
       "      <td>William Henry</td>\n",
       "      <td>Mr</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>8.05000</td>\n",
       "      <td>N</td>\n",
       "      <td>0</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              Age Cabin Embarked     Fare  \\\n",
       "PassengerId                                 \n",
       "1            22.0     N        S   7.2500   \n",
       "2            38.0   C85        C  71.2833   \n",
       "3            26.0     N        S   7.9250   \n",
       "4            35.0  C123        S  53.1000   \n",
       "5            35.0     N        S   8.0500   \n",
       "\n",
       "                                                          Name  Parch  Pclass  \\\n",
       "PassengerId                                                                     \n",
       "1                                      Braund, Mr. Owen Harris      0       3   \n",
       "2            Cumings, Mrs. John Bradley (Florence Briggs Th...      0       1   \n",
       "3                                       Heikkinen, Miss. Laina      0       3   \n",
       "4                 Futrelle, Mrs. Jacques Heath (Lily May Peel)      0       1   \n",
       "5                                     Allen, Mr. William Henry      0       3   \n",
       "\n",
       "                Sex  SibSp  Survived  ...   LastName  \\\n",
       "PassengerId                           ...              \n",
       "1              male      1       0.0  ...     Braund   \n",
       "2            female      1       1.0  ...    Cumings   \n",
       "3            female      0       1.0  ...  Heikkinen   \n",
       "4            female      1       1.0  ...   Futrelle   \n",
       "5              male      0       0.0  ...      Allen   \n",
       "\n",
       "                                         FirstName Designation NewAge  \\\n",
       "PassengerId                                                             \n",
       "1                                      Owen Harris          Mr   22.0   \n",
       "2            John Bradley (Florence Briggs Thayer)         Mrs   38.0   \n",
       "3                                            Laina        Miss   26.0   \n",
       "4                    Jacques Heath (Lily May Peel)         Mrs   35.0   \n",
       "5                                    William Henry          Mr   35.0   \n",
       "\n",
       "             SharedTickets  NormalisedFare  CabinType NoCabin  \\\n",
       "PassengerId                                                     \n",
       "1                        1         7.25000          N       0   \n",
       "2                        2        35.64165          C       1   \n",
       "3                        1         7.92500          N       0   \n",
       "4                        2        26.55000          C       1   \n",
       "5                        1         8.05000          N       0   \n",
       "\n",
       "            GoodSharedTickets  Alone  \n",
       "PassengerId                           \n",
       "1                       False   True  \n",
       "2                        True  False  \n",
       "3                       False   True  \n",
       "4                        True  False  \n",
       "5                       False   True  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combine.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Recall that we had tickets costing more than \\$500. Let's take a closer look at those tickets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Age</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Name</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Survived</th>\n",
       "      <th>...</th>\n",
       "      <th>LastName</th>\n",
       "      <th>FirstName</th>\n",
       "      <th>Designation</th>\n",
       "      <th>NewAge</th>\n",
       "      <th>SharedTickets</th>\n",
       "      <th>NormalisedFare</th>\n",
       "      <th>CabinType</th>\n",
       "      <th>NoCabin</th>\n",
       "      <th>GoodSharedTickets</th>\n",
       "      <th>Alone</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>259</th>\n",
       "      <td>35.0</td>\n",
       "      <td>N</td>\n",
       "      <td>C</td>\n",
       "      <td>512.3292</td>\n",
       "      <td>Ward, Miss. Anna</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Ward</td>\n",
       "      <td>Anna</td>\n",
       "      <td>Miss</td>\n",
       "      <td>35.0</td>\n",
       "      <td>4</td>\n",
       "      <td>128.0823</td>\n",
       "      <td>N</td>\n",
       "      <td>0</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>680</th>\n",
       "      <td>36.0</td>\n",
       "      <td>B51 B53 B55</td>\n",
       "      <td>C</td>\n",
       "      <td>512.3292</td>\n",
       "      <td>Cardeza, Mr. Thomas Drake Martinez</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>male</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Cardeza</td>\n",
       "      <td>Thomas Drake Martinez</td>\n",
       "      <td>Mr</td>\n",
       "      <td>36.0</td>\n",
       "      <td>4</td>\n",
       "      <td>128.0823</td>\n",
       "      <td>B</td>\n",
       "      <td>3</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>738</th>\n",
       "      <td>35.0</td>\n",
       "      <td>B101</td>\n",
       "      <td>C</td>\n",
       "      <td>512.3292</td>\n",
       "      <td>Lesurer, Mr. Gustave J</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>male</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>Lesurer</td>\n",
       "      <td>Gustave J</td>\n",
       "      <td>Mr</td>\n",
       "      <td>35.0</td>\n",
       "      <td>4</td>\n",
       "      <td>128.0823</td>\n",
       "      <td>B</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1235</th>\n",
       "      <td>58.0</td>\n",
       "      <td>B51 B53 B55</td>\n",
       "      <td>C</td>\n",
       "      <td>512.3292</td>\n",
       "      <td>Cardeza, Mrs. James Warburton Martinez (Charlo...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>female</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>Cardeza</td>\n",
       "      <td>James Warburton Martinez (Charlotte Wardle Drake)</td>\n",
       "      <td>Mrs</td>\n",
       "      <td>58.0</td>\n",
       "      <td>4</td>\n",
       "      <td>128.0823</td>\n",
       "      <td>B</td>\n",
       "      <td>3</td>\n",
       "      <td>True</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "              Age        Cabin Embarked      Fare  \\\n",
       "PassengerId                                         \n",
       "259          35.0            N        C  512.3292   \n",
       "680          36.0  B51 B53 B55        C  512.3292   \n",
       "738          35.0         B101        C  512.3292   \n",
       "1235         58.0  B51 B53 B55        C  512.3292   \n",
       "\n",
       "                                                          Name  Parch  Pclass  \\\n",
       "PassengerId                                                                     \n",
       "259                                           Ward, Miss. Anna      0       1   \n",
       "680                         Cardeza, Mr. Thomas Drake Martinez      1       1   \n",
       "738                                     Lesurer, Mr. Gustave J      0       1   \n",
       "1235         Cardeza, Mrs. James Warburton Martinez (Charlo...      1       1   \n",
       "\n",
       "                Sex  SibSp  Survived  ... LastName  \\\n",
       "PassengerId                           ...            \n",
       "259          female      0       1.0  ...     Ward   \n",
       "680            male      0       1.0  ...  Cardeza   \n",
       "738            male      0       1.0  ...  Lesurer   \n",
       "1235         female      0       NaN  ...  Cardeza   \n",
       "\n",
       "                                                     FirstName Designation  \\\n",
       "PassengerId                                                                  \n",
       "259                                                       Anna        Miss   \n",
       "680                                      Thomas Drake Martinez          Mr   \n",
       "738                                                  Gustave J          Mr   \n",
       "1235         James Warburton Martinez (Charlotte Wardle Drake)         Mrs   \n",
       "\n",
       "            NewAge  SharedTickets  NormalisedFare  CabinType NoCabin  \\\n",
       "PassengerId                                                            \n",
       "259           35.0              4        128.0823          N       0   \n",
       "680           36.0              4        128.0823          B       3   \n",
       "738           35.0              4        128.0823          B       1   \n",
       "1235          58.0              4        128.0823          B       3   \n",
       "\n",
       "            GoodSharedTickets  Alone  \n",
       "PassengerId                           \n",
       "259                      True  False  \n",
       "680                      True  False  \n",
       "738                      True  False  \n",
       "1235                     True  False  \n",
       "\n",
       "[4 rows x 21 columns]"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combine[combine.Fare >= 500]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is something similar across all ticket-holders; they all share the same ticket! One theory might be that the ticket fare, \\$512.3292, might reflect the joint total instead of the fare each person paid. If we believe this to be true, we can simply create a new fare column by using the fare to divide by the number of people that shared the same ticket."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "combine['NormalisedFare'] = combine.Fare / combine.SharedTickets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Max Fare is 512.329200\n",
      "Max Normalised Fare is 128.082300\n"
     ]
    }
   ],
   "source": [
    "print('Max Fare is %f' % max(combine.Fare))\n",
    "print('Max Normalised Fare is %f' % max(combine.NormalisedFare))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABIsAAALJCAYAAAA0z6MzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nMyMiAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10accc4e0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Hypothesis 2\n",
    "plt.figure(figsize=(20, 12))\n",
    "sns.distplot(combine[combine.Survived == 0].NormalisedFare, bins=range(0, 131, 1), color = 'red', label='Perished')\n",
    "sns.distplot(combine[combine.Survived == 1].NormalisedFare, bins=range(0, 131, 1), color = 'blue', label='Survived')\n",
    "plt.legend(loc='upper right')\n",
    "plt.title(\"Age Distribution\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "All in all, the observation that passengers who paid higher fares has a higher rate of survival still stands. Now that we have generated some features, we can proceed to develop a Machine Learning algorithm.\n",
    "\n",
    "---\n",
    "\n",
    "#### Investing the Cabin Variable\n",
    "\n",
    "Finally, we investigate the cabin variable, and check whether we are able to generate any new features from it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['A10', 'A11', 'A14', 'A16', 'A18', 'A19', 'A20', 'A21', 'A23',\n",
       "       'A24', 'A26', 'A29', 'A31', 'A32', 'A34', 'A36', 'A5', 'A6', 'A7',\n",
       "       'A9', 'B10', 'B101', 'B102', 'B11', 'B18', 'B19', 'B20', 'B22',\n",
       "       'B24', 'B26', 'B28', 'B3', 'B30', 'B35', 'B36', 'B37', 'B38',\n",
       "       'B39', 'B4', 'B41', 'B42', 'B45', 'B49', 'B5', 'B50',\n",
       "       'B51 B53 B55', 'B52 B54 B56', 'B57 B59 B63 B66', 'B58 B60', 'B61',\n",
       "       'B69', 'B71', 'B73', 'B77', 'B78', 'B79', 'B80', 'B82 B84', 'B86',\n",
       "       'B94', 'B96 B98', 'C101', 'C103', 'C104', 'C105', 'C106', 'C110',\n",
       "       'C111', 'C116', 'C118', 'C123', 'C124', 'C125', 'C126', 'C128',\n",
       "       'C130', 'C132', 'C148', 'C2', 'C22 C26', 'C23 C25 C27', 'C28',\n",
       "       'C30', 'C31', 'C32', 'C39', 'C45', 'C46', 'C47', 'C49', 'C50',\n",
       "       'C51', 'C52', 'C53', 'C54', 'C55 C57', 'C6', 'C62 C64', 'C65',\n",
       "       'C68', 'C7', 'C70', 'C78', 'C80', 'C82', 'C83', 'C85', 'C86',\n",
       "       'C87', 'C89', 'C90', 'C91', 'C92', 'C93', 'C95', 'C97', 'C99', 'D',\n",
       "       'D10 D12', 'D11', 'D15', 'D17', 'D19', 'D20', 'D21', 'D22', 'D26',\n",
       "       'D28', 'D30', 'D33', 'D34', 'D35', 'D36', 'D37', 'D38', 'D40',\n",
       "       'D43', 'D45', 'D46', 'D47', 'D48', 'D49', 'D50', 'D56', 'D6', 'D7',\n",
       "       'D9', 'E10', 'E101', 'E12', 'E121', 'E17', 'E24', 'E25', 'E31',\n",
       "       'E33', 'E34', 'E36', 'E38', 'E39 E41', 'E40', 'E44', 'E45', 'E46',\n",
       "       'E49', 'E50', 'E52', 'E58', 'E60', 'E63', 'E67', 'E68', 'E77',\n",
       "       'E8', 'F', 'F E46', 'F E57', 'F E69', 'F G63', 'F G73', 'F2',\n",
       "       'F33', 'F38', 'F4', 'G6', 'N', 'T'], dtype=object)"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.unique(combine.Cabin)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We note that there are many different types of cabins, and some passengers \"share\" cabins with other passengers. To capture these effects, we can create a new column, \"CabinType\", using the first character of the Cabin feature, and another, \"NoCabin\" to capture the number of cabins."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/ngbolin/anaconda3/lib/python3.6/site-packages/seaborn/categorical.py:3666: UserWarning: The `factorplot` function has been renamed to `catplot`. The original name will be removed in a future release. Please update your code. Note that the default `kind` in `factorplot` (`'point'`) has changed `'strip'` in `catplot`.\n",
      "  warnings.warn(msg)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABJIAAAJQCAYAAADVHU6aAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10ab80dd8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "combine['CabinType'] = combine.Cabin.apply(lambda x: x[0])\n",
    "\n",
    "def number_of_cabins(cabin):\n",
    "    if cabin == 'N': return 0\n",
    "    else: return len(cabin.split(' '))\n",
    "    \n",
    "combine['NoCabin'] = combine.Cabin.apply(lambda x: number_of_cabins(x))\n",
    "\n",
    "fig = plt.figure(figsize=(20, 10))\n",
    "ax1 = fig.add_subplot(121); ax2 = fig.add_subplot(122)\n",
    "\n",
    "g1 = sns.factorplot('NoCabin', 'Survived', \n",
    "                    kind = 'bar', data = combine, ax = ax1)\n",
    "plt.close(g1.fig)\n",
    "g2 = sns.factorplot('CabinType', 'Survived', \n",
    "                    kind = 'bar', data = combine, ax = ax2)\n",
    "plt.close(g2.fig)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As it turns out, passengers staying in cabins had a higher possibility of surviving, but the effect does not increase with the number of cabins. In fact, the higher number of cabins, the lower the probability of the passenger surviving. We can then convert this column to categorical instead of numerical to capture the asymmetric effects.\n",
    "\n",
    "From the Cabin types, we note that passengers staying in the \"E\", \"D\", and \"B\" cabins had a higher survival rate. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "category\n"
     ]
    }
   ],
   "source": [
    "# Not very important to do\n",
    "combine['NoCabin'] = pd.Categorical(combine.NoCabin)\n",
    "print(combine.NoCabin.dtypes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Selection\n",
    "\n",
    "After all this, we can begin to select features which we believe are informative of the response variable. We typically do so using a heatmap/correlation plot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABGoAAAMGCAYAAACqJmKpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nzMwKxP8DiA4BXmtmnqkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1090d4f28>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(20, 12))\n",
    "sns.heatmap(combine.corr(), cmap='Blues', annot=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the correlation plot, we note that the following features seem to be fairly highly correlated with the \"Survived\" feature:\n",
    "\n",
    "1. NormalisedFare\n",
    "2. GoodSharedTickets\n",
    "3. NewAge\n",
    "4. PassengerClass\n",
    "\n",
    "In addition, we will include some other categorical features not included in the correlation plots:\n",
    "\n",
    "1. Gender\n",
    "2. CabinType\n",
    "3. NoCabin\n",
    "4. Designation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "features = ['NormalisedFare', 'GoodSharedTickets', 'NewAge', 'Pclass', 'Alone',\n",
    "            'Sex', 'CabinType', 'NoCabin', 'Designation']\n",
    "\n",
    "X = combine[features]\n",
    "y = combine['Survived']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Developing a Machine Learning Algorithm\n",
    "\n",
    "Noting that the problem we are currently dealing with is a categorical supervised learning problem, we can use the following techniques for prediction:\n",
    "\n",
    "1. Random Forest\n",
    "2. K-Nearest Neighbors\n",
    "\n",
    "#### Training, Cross-Validation and Testing\n",
    "Before we begin developing the models, it is important to split our training data into the following:\n",
    "\n",
    "1. Training data\n",
    "2. Cross-validation data\n",
    "\n",
    "The training data will be used solely for the purpose of training our classifier, while the cross-validation data will be used to evaluate the performance of the model. Typically, the training-validation data split is 80-20, but depending on data availability and problem scope, it is appropriate to use a 70-30 or 90-10 split. For our case, we will be using a 90-10 split, as we are dealing with only 891 observations.\n",
    "\n",
    "We will be using the performance on the cross-validation dataset as a metric for prediction on the test dataset. Why would we want to do that? As it turns out, as we increase the number of parameters in our model, our model becomes so complicated that it cannot be justified by the data (the model is **\"memorizing\"** the observations), and cannot generalise well to unseen data.\n",
    "\n",
    "As a result, splitting the data into training and cross-validation helps to reduce the problem of overfitting, and helps improve generalizability and performance on previously unseen dataset.\n",
    "\n",
    "---\n",
    "\n",
    "##### Model 1: Random Forest Classifier\n",
    "Recall that a random forest classifier is an ensemble of decision trees. This begs the question: What's a decision tree? Well, a decision tree is a classifier that makes decisions using decision rules. For example, suppose we are trying to predict whether or not someone is going to pass the final exams or not; we can use decision rules such as \"If the subject scored more than 70 points for his mid-term\", he is likely to pass the exam. Of course, this posits that we have data on his midterm. Then, the random forest classifier collects all these decision trees, with each decision tree splits on different features (based on what the classifier believes to be the best feature to split on). Then it passes the validation data through all these trees, and collects the result.\n",
    "\n",
    "With that in mind, we can start creating our first Machine Learning model - a Random Forest classifier. We can easily do so in Python, using the following function: `from sklearn.ensemble import RandomForestClassifier`.\n",
    "\n",
    "The classifier (you can find more information [here](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html)) takes the following (and more) as inputs:\n",
    "\n",
    "1. n_estimators - the number of decision trees, typically set to 500 or 1000\n",
    "2. criterion - the function that measures the quality of the split to decide which variable/feature to split on\n",
    "3. max_depth - maximum depth of the decision tree\n",
    "4. max_features - the number of maximum features to consider when looking for the best split\n",
    "5. min_samples_split - the minimum number of samples required to split an internal node\n",
    "6. random_state - we set a random_state to ensure reproducibility\n",
    "\n",
    "For our classifier, we will also be using the GridSearch technique - we iterate over possible hyperparameters and look for the best combination of parameters that yields the highest accuracy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/ngbolin/anaconda3/lib/python3.6/site-packages/ipykernel_launcher.py:12: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  if sys.path[0] == '':\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "def label_coding(col):\n",
    "    encoder = LabelEncoder()\n",
    "    return encoder.fit_transform(col)\n",
    "\n",
    "categorical_columns = ['Designation', 'Sex', 'CabinType']\n",
    "for col in categorical_columns:\n",
    "    newcolname = str(col)\n",
    "    X[newcolname] = label_coding(X[col])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "891"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Splitting Training Data\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "random_state = 42\n",
    "X_train = X[:len(train)].copy(); y_train = y[:len(train)].copy()\n",
    "X_test = X[len(train):].copy()\n",
    "\n",
    "# Splitting into training and cross-validation dataset\n",
    "X_subtrain, X_subtest, y_subtrain, y_subtest = train_test_split(X_train, y_train, \n",
    "                                                                test_size = 0.2,\n",
    "                                                                random_state = random_state)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8329510395707578\n"
     ]
    }
   ],
   "source": [
    "# Model 1: Random Forest Classifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "def random_forest(X_subtrain, y_subtrain):\n",
    "    # Defining the classifier and its arguments\n",
    "    rf = RandomForestClassifier(n_estimators = 1000,\n",
    "                                criterion = 'gini',\n",
    "                                min_samples_split=10,\n",
    "                                random_state = random_state)\n",
    "    \n",
    "    param_grid = {'max_depth': [4, 5, 6]}\n",
    "    rf_clf = GridSearchCV(rf, param_grid).fit(X_subtrain, y_subtrain)\n",
    "    rf_score = cross_val_score(rf_clf, X_subtrain, y_subtrain, cv = 10).mean()\n",
    "    return rf_clf, rf_score\n",
    "\n",
    "rf_clf, rf_score = random_forest(X_subtrain, y_subtrain)\n",
    "print(rf_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using cross-validation, we achieved an accuracy of 83.3% over 10-fold cross validation. What goes on under the hood?\n",
    "\n",
    "<img src=\"images/cross-validation.png\">\n",
    "\n",
    "Python splits up the dataset into 10 different folds, and uses 9 of them for training the classifier. It then evaluates the classifier's performance on the tenth fold. It does this 10 different times, each time using a different fold for testing. The accuracy score we see is the average (mean) accuracy across 10 different iterations. This begs the question: How well does our model perform? To answer this, we can simply check the average survival rate in the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.3838383838383838"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(combine.Survived)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As it turns out, a simple model that predicts 0 (everyone perishes) for every observation would yield an accuracy score of 61.6%. However, our model had an accuracy of 83.3%, beating the simple baseline model by more than 20%.\n",
    "\n",
    "---\n",
    "\n",
    "##### Model 2: K-Nearest Neighbors\n",
    "\n",
    "The next model that we are interested in is the k-Nearest Neighbors model. What exactly is it? The k-Nearest Neighbors algorithm is a non-parametric method that takes an observation as its input, and returns the output from the k different observations that are the most similar to (smallest distance from) the observation.\n",
    "\n",
    "How do we judge distance? One way to do so is through the Euclidean distance metric. But what if there is reason to believe that for some categorical variables, there is no natural ordering i.e. Low - Medium - High? There are ways to circumvent this issue, but we will not be discussing those issues here. Instead, we sweep these questions under the rug (you can learn more about it [here](https://www.quora.com/How-can-I-use-KNN-for-mixed-data-categorical-and-numerical)), and proceed to encode our categorical variables.\n",
    "\n",
    "To proceed with the algorithm, here are some steps we need to take:\n",
    "\n",
    "1. Import the library and the corresponding functions that allow us to use the algorithm.\n",
    "2. Choose K - typically, K is chosen to be an odd number, such as 5 or 7. We will use 5 in our case.\n",
    "3. We scale our datasets; the K-Nearest Neighbors algorithm works best with scaled data.\n",
    "4. We pass our data into the algorithm, and use that to predict for the validation data and verify it with the actual realisations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.7261323496534764\n"
     ]
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "def k_nearest_neighbors(X_subtrain, y_subtrain):\n",
    "    # Scaling the data\n",
    "    scaler = StandardScaler()\n",
    "    X_train = scaler.fit_transform(X_subtrain)\n",
    "    \n",
    "    knn = KNeighborsClassifier()\n",
    "    param_grid = {'n_neighbors': [3, 5, 7, 10],\n",
    "                  'weights': ['uniform', 'distance']}\n",
    "    knn_clf = GridSearchCV(knn, param_grid).fit(X_subtrain, y_subtrain)\n",
    "    knn_score = cross_val_score(knn_clf, X_subtrain, y_subtrain, cv = 10).mean()\n",
    "    return knn_clf, knn_score\n",
    "\n",
    "knn_clf, knn_score = k_nearest_neighbors(X_subtrain, y_subtrain)\n",
    "print(knn_score)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It appears that the KNN model did not perform as well as the Random Forest classifier, since it only achieved an accuracy of 72.6%. Nonetheless, let's check the classifiers' performance on the validation dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.8156424581005587\n",
      "0.7374301675977654\n"
     ]
    }
   ],
   "source": [
    "print(rf_clf.score(X_subtest, y_subtest))\n",
    "print(knn_clf.score(X_subtest, y_subtest))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It turns out that our intuition was right - the Random Forest classifier did perform better on the validation dataset. \n",
    "\n",
    "Let's use the Random Forest classifier to predict for the testing dataset, and submit it to Kaggle and check how well we did."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [],
   "source": [
    "submission.to_csv('submission1.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def submissions():\n",
    "    submission = pd.read_csv('data/titanic/gender_submission.csv')\n",
    "    submission['Survived'] = [int(x) for x in rf_clf.predict(X_test)]\n",
    "    submission.to_csv('data/titanic/submissions.csv', index=False)\n",
    "    \n",
    "submissions()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our model did pretty well! Please feel free to improve on our baseline model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}