SushilShrestha/Quantitative Analysis of Apps on Google Playstore.ipynb

## Quantitative Analysis of Apps on Google Playstore.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analysis of Google Playstore Apps\n",
    "**Sushil Shrestha (sushilsh@hawaii.edu)**<br />\n",
    "**ICS 691D final project**\n",
    "______\n",
    "\n",
    "# Project Overview\n",
    "The Google Play store is the most popular and widely used android app market place. It has above 2.5 millions android apps presently available for browse and download[$^{[1]}$](https://www.statista.com/statistics/266210/number-of-available-applications-in-the-google-play-store/). The most popular app on the Play store are `Google Play services`, `YouTube` and `Google Maps`. Beside these `Facebook` is most downloaded third party app hosted in the Google Play store[$^{[2]}$](https://en.wikipedia.org/wiki/List_of_most-downloaded_Google_Play_applications#Free_applications_with_over_five_billion_downloads).\n",
    "\n",
    "For this project, we'll look at the application data in the Google Play store. The data set we are using for the analysis is from Kaggle [Google Play store apps dataset](https://www.kaggle.com/lava18/google-play-store-apps). The dataset contains around 10k entries and 13 features. As initial step for our analysis, we will clean our data and impute our missing values using mean and a linear model. We will be discussing about following research questions.\n",
    "\n",
    "**Research Questions?**\n",
    "    1. Is price of apps on Google Play store significantly different than price of apps on Apple App store?\n",
    "    2. Is the quality of paid apps on Google Play store significantly different than free apps?\n",
    "    3. Can we build a classifier to classify the apps to its category?\n",
    "\n",
    "# Methods\n",
    "Let's begin our analysis by importing libraries we'll require for our analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import scipy.stats as stats\n",
    "import seaborn as sns\n",
    "from sklearn import preprocessing\n",
    "import warnings\n",
    "\n",
    "warnings.filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's read our dataset for the google playstore and change the column names to upper case."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>APP</th>\n",
       "      <th>CATEGORY</th>\n",
       "      <th>RATING</th>\n",
       "      <th>REVIEWS</th>\n",
       "      <th>SIZE</th>\n",
       "      <th>INSTALLS</th>\n",
       "      <th>TYPE</th>\n",
       "      <th>PRICE</th>\n",
       "      <th>CONTENT RATING</th>\n",
       "      <th>GENRES</th>\n",
       "      <th>LAST UPDATED</th>\n",
       "      <th>CURRENT VER</th>\n",
       "      <th>ANDROID VER</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Photo Editor &amp; Candy Camera &amp; Grid &amp; ScrapBook</td>\n",
       "      <td>ART_AND_DESIGN</td>\n",
       "      <td>4.1</td>\n",
       "      <td>159</td>\n",
       "      <td>19M</td>\n",
       "      <td>10,000+</td>\n",
       "      <td>Free</td>\n",
       "      <td>0</td>\n",
       "      <td>Everyone</td>\n",
       "      <td>Art &amp; Design</td>\n",
       "      <td>January 7, 2018</td>\n",
       "      <td>1.0.0</td>\n",
       "      <td>4.0.3 and up</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Coloring book moana</td>\n",
       "      <td>ART_AND_DESIGN</td>\n",
       "      <td>3.9</td>\n",
       "      <td>967</td>\n",
       "      <td>14M</td>\n",
       "      <td>500,000+</td>\n",
       "      <td>Free</td>\n",
       "      <td>0</td>\n",
       "      <td>Everyone</td>\n",
       "      <td>Art &amp; Design;Pretend Play</td>\n",
       "      <td>January 15, 2018</td>\n",
       "      <td>2.0.0</td>\n",
       "      <td>4.0.3 and up</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              APP        CATEGORY  RATING  \\\n",
       "0  Photo Editor & Candy Camera & Grid & ScrapBook  ART_AND_DESIGN     4.1   \n",
       "1                             Coloring book moana  ART_AND_DESIGN     3.9   \n",
       "\n",
       "  REVIEWS SIZE  INSTALLS  TYPE PRICE CONTENT RATING  \\\n",
       "0     159  19M   10,000+  Free     0       Everyone   \n",
       "1     967  14M  500,000+  Free     0       Everyone   \n",
       "\n",
       "                      GENRES      LAST UPDATED CURRENT VER   ANDROID VER  \n",
       "0               Art & Design   January 7, 2018       1.0.0  4.0.3 and up  \n",
       "1  Art & Design;Pretend Play  January 15, 2018       2.0.0  4.0.3 and up  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "playstore_app_data = pd.read_csv(\"google-play-store-apps/googleplaystore.csv\")\n",
    "playstore_app_data.columns = playstore_app_data.columns.str.upper()\n",
    "playstore_app_data.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have 13 columns, 7 of them seems to be categorical, 5 numerical and 1 date. Next step would be to start cleaning our data for our analysis."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Cleaning\n",
    "### Misaligned data\n",
    "During manual inspection of the data entries, one of the data row was found to be misaliged, so the misaligned entry was recorded and data was aligned as required."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>APP</th>\n",
       "      <th>CATEGORY</th>\n",
       "      <th>RATING</th>\n",
       "      <th>REVIEWS</th>\n",
       "      <th>SIZE</th>\n",
       "      <th>INSTALLS</th>\n",
       "      <th>TYPE</th>\n",
       "      <th>PRICE</th>\n",
       "      <th>CONTENT RATING</th>\n",
       "      <th>GENRES</th>\n",
       "      <th>LAST UPDATED</th>\n",
       "      <th>CURRENT VER</th>\n",
       "      <th>ANDROID VER</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10472</th>\n",
       "      <td>Life Made WI-Fi Touchscreen Photo Frame</td>\n",
       "      <td>1.9</td>\n",
       "      <td>19.0</td>\n",
       "      <td>3.0M</td>\n",
       "      <td>1,000+</td>\n",
       "      <td>Free</td>\n",
       "      <td>0</td>\n",
       "      <td>Everyone</td>\n",
       "      <td>NaN</td>\n",
       "      <td>February 11, 2018</td>\n",
       "      <td>1.0.19</td>\n",
       "      <td>4.0 and up</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           APP CATEGORY  RATING REVIEWS  \\\n",
       "10472  Life Made WI-Fi Touchscreen Photo Frame      1.9    19.0    3.0M   \n",
       "\n",
       "         SIZE INSTALLS TYPE     PRICE CONTENT RATING             GENRES  \\\n",
       "10472  1,000+     Free    0  Everyone            NaN  February 11, 2018   \n",
       "\n",
       "      LAST UPDATED CURRENT VER ANDROID VER  \n",
       "10472       1.0.19  4.0 and up         NaN  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "playstore_app_data[playstore_app_data['PRICE'] == 'Everyone']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>APP</th>\n",
       "      <th>CATEGORY</th>\n",
       "      <th>RATING</th>\n",
       "      <th>REVIEWS</th>\n",
       "      <th>SIZE</th>\n",
       "      <th>INSTALLS</th>\n",
       "      <th>TYPE</th>\n",
       "      <th>PRICE</th>\n",
       "      <th>CONTENT RATING</th>\n",
       "      <th>GENRES</th>\n",
       "      <th>LAST UPDATED</th>\n",
       "      <th>CURRENT VER</th>\n",
       "      <th>ANDROID VER</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10472</th>\n",
       "      <td>Life Made WI-Fi Touchscreen Photo Frame</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.9</td>\n",
       "      <td>19.0</td>\n",
       "      <td>3.0M</td>\n",
       "      <td>1,000+</td>\n",
       "      <td>Free</td>\n",
       "      <td>0</td>\n",
       "      <td>Everyone</td>\n",
       "      <td>NaN</td>\n",
       "      <td>February 11, 2018</td>\n",
       "      <td>1.0.19</td>\n",
       "      <td>4.0 and up</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           APP  CATEGORY  RATING  REVIEWS  \\\n",
       "10472  Life Made WI-Fi Touchscreen Photo Frame       NaN     1.9     19.0   \n",
       "\n",
       "       SIZE INSTALLS  TYPE PRICE CONTENT RATING GENRES       LAST UPDATED  \\\n",
       "10472  3.0M   1,000+  Free     0       Everyone    NaN  February 11, 2018   \n",
       "\n",
       "      CURRENT VER ANDROID VER  \n",
       "10472      1.0.19  4.0 and up  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "misaligned_data = playstore_app_data[playstore_app_data['PRICE'] == 'Everyone']\n",
    "misaligned_data = misaligned_data.shift(periods=1, axis='columns')\n",
    "misaligned_data.APP = misaligned_data.CATEGORY\n",
    "misaligned_data.CATEGORY = np.NaN\n",
    "misaligned_data.RATING = 1.9\n",
    "misaligned_data.REVIEWS = 19.0\n",
    "\n",
    "playstore_app_data.loc[playstore_app_data['PRICE'] == 'Everyone', :] = misaligned_data\n",
    "misaligned_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Type casting\n",
    "As we have noted, there are 5 numerical fields in our dataset. Let look at the how they are represented in out data frame and cast them to numerical data type if required."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 10841 entries, 0 to 10840\n",
      "Data columns (total 13 columns):\n",
      "APP               10841 non-null object\n",
      "CATEGORY          10840 non-null object\n",
      "RATING            9367 non-null float64\n",
      "REVIEWS           10841 non-null object\n",
      "SIZE              10841 non-null object\n",
      "INSTALLS          10841 non-null object\n",
      "TYPE              10840 non-null object\n",
      "PRICE             10841 non-null object\n",
      "CONTENT RATING    10841 non-null object\n",
      "GENRES            10840 non-null object\n",
      "LAST UPDATED      10841 non-null object\n",
      "CURRENT VER       10833 non-null object\n",
      "ANDROID VER       10839 non-null object\n",
      "dtypes: float64(1), object(12)\n",
      "memory usage: 1.1+ MB\n"
     ]
    }
   ],
   "source": [
    "playstore_app_data.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`REVIEWS`, `SIZE`, `INSTALLS` and `PRICE` are represented as string object in our dataframe. We will have to cast `REVIEWS` to `int`, `SIZE` field to `float`, `INSTALLS` to `float` and `PRICE` field to `float`, that will make our 5 numerical fields ready for analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>APP</th>\n",
       "      <th>CATEGORY</th>\n",
       "      <th>RATING</th>\n",
       "      <th>REVIEWS</th>\n",
       "      <th>SIZE</th>\n",
       "      <th>INSTALLS</th>\n",
       "      <th>TYPE</th>\n",
       "      <th>PRICE</th>\n",
       "      <th>CONTENT RATING</th>\n",
       "      <th>GENRES</th>\n",
       "      <th>LAST UPDATED</th>\n",
       "      <th>CURRENT VER</th>\n",
       "      <th>ANDROID VER</th>\n",
       "      <th>SIZE_NUM</th>\n",
       "      <th>INSTALLS_NUM</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Photo Editor &amp; Candy Camera &amp; Grid &amp; ScrapBook</td>\n",
       "      <td>ART_AND_DESIGN</td>\n",
       "      <td>4.1</td>\n",
       "      <td>159</td>\n",
       "      <td>19M</td>\n",
       "      <td>10,000+</td>\n",
       "      <td>Free</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Everyone</td>\n",
       "      <td>Art &amp; Design</td>\n",
       "      <td>January 7, 2018</td>\n",
       "      <td>1.0.0</td>\n",
       "      <td>4.0.3 and up</td>\n",
       "      <td>19.0</td>\n",
       "      <td>10000.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              APP        CATEGORY  RATING  \\\n",
       "0  Photo Editor & Candy Camera & Grid & ScrapBook  ART_AND_DESIGN     4.1   \n",
       "\n",
       "   REVIEWS SIZE INSTALLS  TYPE  PRICE CONTENT RATING        GENRES  \\\n",
       "0      159  19M  10,000+  Free    0.0       Everyone  Art & Design   \n",
       "\n",
       "      LAST UPDATED CURRENT VER   ANDROID VER  SIZE_NUM  INSTALLS_NUM  \n",
       "0  January 7, 2018       1.0.0  4.0.3 and up      19.0       10000.0  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "def app_size_to_num(appsize):\n",
    "    '''\n",
    "    convert 12M => 12 and 12k => 12/1024\n",
    "    '''\n",
    "    if appsize == 'Varies with device':\n",
    "        return np.nan\n",
    "    MB_match = re.match(r'(\\d+(?:\\.\\d+)?)M', appsize)\n",
    "    KB_match = re.match(r'(\\d+(?:\\.\\d+)?)k', appsize)\n",
    "    if MB_match:\n",
    "        return float(MB_match.group(1))\n",
    "    elif KB_match:\n",
    "        return float(KB_match.group(1))/1024\n",
    "    return np.nan\n",
    "\n",
    "playstore_app_data['SIZE_NUM'] = playstore_app_data.SIZE.map(app_size_to_num)\n",
    "playstore_app_data['INSTALLS_NUM'] = playstore_app_data.INSTALLS.str.replace(',', '').str.replace('+', '').astype(float)\n",
    "playstore_app_data['REVIEWS'] = playstore_app_data.REVIEWS.astype(int)\n",
    "playstore_app_data['PRICE'] = playstore_app_data['PRICE'].replace('[\\$,]', '', regex=True).astype(float)\n",
    "\n",
    "playstore_app_data_dropna = playstore_app_data.dropna()\n",
    "\n",
    "playstore_app_data.head(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Impute missing values\n",
    "Another important step for the analysis is dealing with the missing values. The original data may have missing values and that might break our algorithms later in the analysis. Lets look at which fields have major missing values and fix them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "APP                  0\n",
       "CATEGORY             1\n",
       "RATING            1474\n",
       "REVIEWS              0\n",
       "SIZE                 0\n",
       "INSTALLS             0\n",
       "TYPE                 1\n",
       "PRICE                0\n",
       "CONTENT RATING       0\n",
       "GENRES               1\n",
       "LAST UPDATED         0\n",
       "CURRENT VER          8\n",
       "ANDROID VER          2\n",
       "SIZE_NUM          1695\n",
       "INSTALLS_NUM         0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "playstore_app_data.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It seems like `SIZE_NUM` and `REVIEWS` columns have majority of missing values. As we have studied, one technique to handle missing values is to replace the missing values with the mean of the column. In our case, replacing the missing `SIZE_NUM` by the mean of the size of app in that category makes more sense so lets replace the missing `SIZE_NUM` by mean size of apps on that category."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def size_imputer(row):\n",
    "    if not pd.np.isnan(row.SIZE_NUM):\n",
    "        return row\n",
    "    df_category = playstore_app_data[playstore_app_data['CATEGORY'] == row['CATEGORY']]\n",
    "    mean_size = np.mean(df_category['SIZE_NUM'].dropna())\n",
    "    row['SIZE_NUM'] = mean_size\n",
    "    return row\n",
    "\n",
    "playstore_app_data_imputed = playstore_app_data.apply(size_imputer, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Relation between Popularity of app and its Size, Rating and Reviews\n",
    "\n",
    "Another field with missing value is the `RATING` field. We would like to try new method to fill in the missing value for `RATING` field. We will first train a linear model and will use that model to predict the `RATING`. Lets first look at how other variables affect the ratings and try to find the dependent variables for our linear model. \n",
    "\n",
    "The Google Playstore dataset has `INSTALLS` field that records the number of times app had been installed. The `INSTALLS` field is a measure of how popular is that app. Now let's look at relation between other field and the popularity of the app. Now lets find the mean size of app, mean number of reviews and mean ratings of apps and compare them with `INSTALLS`. We will take `log` of `INSTALLS` to make data more readable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Number of installs(log scale)')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 792x1152 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mean_app_data = playstore_app_data.dropna().groupby('INSTALLS_NUM').mean().sort_index()\n",
    "\n",
    "plt.figure(figsize=(11, 16))\n",
    "plt.subplot(3, 1, 1)\n",
    "plt.plot(np.log10(mean_app_data.index), mean_app_data['RATING'], label='Mean App Rating')\n",
    "plt.ylabel(\"Mean ratings of apps\")\n",
    "plt.title(\"Ratings, Reviews, Size vs Num of Installs\")\n",
    "plt.subplot(3, 1, 2)\n",
    "plt.plot(np.log10(mean_app_data.index), mean_app_data['REVIEWS'], label='Mean Number of Reviews')\n",
    "plt.ylabel(\"Mean number of reviews\")\n",
    "plt.subplot(3, 1, 3)\n",
    "plt.plot(np.log10(mean_app_data.index), mean_app_data['SIZE_NUM'])\n",
    "plt.ylabel(\"Mean size of apps\")\n",
    "plt.xlabel(\"Number of installs(log scale)\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see as `INSTALLS` increases, the mean `SIZE` of the app increases with few exceptional spikes. One reason for the trend might be that as number of user increases developer can start focusing on providing better experience and facilities than to worry about the app size.\n",
    "\n",
    "For number of `REVIEWS`, we can see a constant plateau curve and a steep rise in number of reviews as size exceeds 10 million+ installs. \n",
    "\n",
    "The average `RATING` seems to be complex than `SIZE` and `REVIEWS`. The average `RATING` of less popular app seems to be high and seems to decline until apps have around 1000+ installs. Then `RATING` seem to rise slightly after app have 100k+ installs."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Correlation Between the variables\n",
    "We can find the correlation between features and have a glance of how they are related with each other. Correlation analysis can be used to find important features that might affect the given variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": false
   },
   "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>RATING</th>\n",
       "      <th>REVIEWS</th>\n",
       "      <th>PRICE</th>\n",
       "      <th>SIZE_NUM</th>\n",
       "      <th>INSTALLS_NUM</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>RATING</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.079819</td>\n",
       "      <td>-0.021320</td>\n",
       "      <td>0.083643</td>\n",
       "      <td>0.052693</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>REVIEWS</th>\n",
       "      <td>0.079819</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.010184</td>\n",
       "      <td>0.240381</td>\n",
       "      <td>0.626187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PRICE</th>\n",
       "      <td>-0.021320</td>\n",
       "      <td>-0.010184</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.026274</td>\n",
       "      <td>-0.010852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SIZE_NUM</th>\n",
       "      <td>0.083643</td>\n",
       "      <td>0.240381</td>\n",
       "      <td>-0.026274</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.162707</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>INSTALLS_NUM</th>\n",
       "      <td>0.052693</td>\n",
       "      <td>0.626187</td>\n",
       "      <td>-0.010852</td>\n",
       "      <td>0.162707</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                RATING   REVIEWS     PRICE  SIZE_NUM  INSTALLS_NUM\n",
       "RATING        1.000000  0.079819 -0.021320  0.083643      0.052693\n",
       "REVIEWS       0.079819  1.000000 -0.010184  0.240381      0.626187\n",
       "PRICE        -0.021320 -0.010184  1.000000 -0.026274     -0.010852\n",
       "SIZE_NUM      0.083643  0.240381 -0.026274  1.000000      0.162707\n",
       "INSTALLS_NUM  0.052693  0.626187 -0.010852  0.162707      1.000000"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "playstore_app_data_dropna.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Correlation Heatmap')"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 792x648 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(style=\"white\")\n",
    "# Compute the correlation matrix\n",
    "corr = playstore_app_data.corr()\n",
    "\n",
    "# Generate a mask for the upper triangle\n",
    "mask = np.zeros_like(corr, dtype=np.bool)\n",
    "mask[np.triu_indices_from(mask)] = True\n",
    "\n",
    "# Set up the matplotlib figure\n",
    "f, ax = plt.subplots(figsize=(11, 9))\n",
    "\n",
    "# Generate a custom diverging colormap\n",
    "cmap = sns.diverging_palette(220, 10, as_cmap=True)\n",
    "\n",
    "# Draw the heatmap with the mask and correct aspect ratio\n",
    "sns.heatmap(corr, mask=mask, cmap=cmap, vmax=.3, center=0,\n",
    "            square=True, linewidths=.5, cbar_kws={\"shrink\": .5})\n",
    "plt.title(\"Correlation Heatmap\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The correlation heatmap shows that there is correlation between `RATING` and other features `INSTALLS_NUM`, `SIZE_NUM` and `REVIEWS`. The `REVIEWS` column is highly correlated with `INSTALLS_NUM`, `SIZE_NUM` AND `RATING`. However `PRICE` feature doesn't seem to correlate with any other features. \n",
    "\n",
    "Since `RATING` field shows correlations with other fields, we will train a `SGDRegressor` to predict the `RATING` field given other features as an input. \n",
    "\n",
    "First of all since our features are widely spread we will normalized the features before we feed them to the `SGDRegressor` model. `INSTALLS_NUM` and `REVIEWS` are log normalized. To prevent `inf` while we take log we will add 1 to the `INSTALLS_NUM` and `REVIEWS` fields before we take log. We'll do the mean normalization for the `PRICE` field.\n",
    "\n",
    "We are using 5 fold cross validation with GridSearch to tune our hyper parameters. The best hypermeters will be then used to calculate the final score of the model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.11084496855342463"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import SGDRegressor\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_data = playstore_app_data_dropna[['REVIEWS', 'PRICE', 'SIZE_NUM', 'INSTALLS_NUM']]\n",
    "y_data = playstore_app_data_dropna['RATING']\n",
    "\n",
    "mean_price = np.mean(X_data['PRICE'])\n",
    "std_price = np.std(X_data['PRICE'])\n",
    "\n",
    "def normalize_data_for_rating(df_data):\n",
    "    df_data['INSTALLS_NUM'] = np.log10(df_data['INSTALLS_NUM']+1)\n",
    "    df_data['REVIEWS'] = np.log10(df_data['REVIEWS']+1)\n",
    "    df_data['PRICE'] = (df_data['PRICE'] - mean_price)/std_price\n",
    "    return df_data\n",
    "\n",
    "X_data = normalize_data_for_rating(X_data)\n",
    "X_train, X_test, y_train, y_test = train_test_split(X_data, y_data, test_size=0.2, random_state=2)\n",
    "\n",
    "parameters = dict(loss=['squared_loss', 'huber', 'epsilon_insensitive', 'squared_epsilon_insensitive'],\n",
    "                  penalty=[ 'l1', 'elasticnet', 'l2'],\n",
    "                  alpha=[0.0001, 0.003, 0.001, 0.03, 0.01]\n",
    "                 )\n",
    "sgdmodel = SGDRegressor(max_iter=1000, tol=1e-3)\n",
    "rating_predictor = GridSearchCV(sgdmodel, parameters, cv=5)\n",
    "rating_predictor.fit(X_train, y_train)\n",
    "\n",
    "rating_predictor.best_params_, rating_predictor.best_score_\n",
    "rating_predictor.score(X_test, y_test)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we have reasonable ratings predictor with output loss of around 0.105. We can use this model to predict our missing `RATING` on our data. So lets use our `SGDRegressor` to impute the missing values in our original data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def ratings_imputer(row):\n",
    "    if not pd.np.isnan(row.RATING):\n",
    "        return row\n",
    "    X_row = pd.DataFrame(row[['REVIEWS', 'PRICE', 'SIZE_NUM', 'INSTALLS_NUM']])\n",
    "    normalized_X = normalize_data_for_rating(X_row.T.astype('float'))\n",
    "    y = rating_predictor.predict(X=normalized_X)\n",
    "    row['RATING'] = y\n",
    "    return row\n",
    "\n",
    "playstore_app_data_imputed = playstore_app_data_imputed.apply(ratings_imputer, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "APP               0\n",
       "CATEGORY          1\n",
       "RATING            0\n",
       "REVIEWS           0\n",
       "SIZE              0\n",
       "INSTALLS          0\n",
       "TYPE              1\n",
       "PRICE             0\n",
       "CONTENT RATING    0\n",
       "GENRES            1\n",
       "LAST UPDATED      0\n",
       "CURRENT VER       8\n",
       "ANDROID VER       2\n",
       "SIZE_NUM          0\n",
       "INSTALLS_NUM      0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "playstore_app_data_imputed.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Other missing features are in minority so we can drop the minority missing data rows."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "playstore_app_data_dropna = playstore_app_data.dropna()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analysis of price of an app on Playstore"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lets see how the price is distributed in the Playstore. We will begin with average price of an app on Google Playstore then see the KDE plot for more detail view."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average price of an app in Google Playstore is  1.0272733142699015\n"
     ]
    }
   ],
   "source": [
    "print(\"Average price of an app in Google Playstore is \", np.average(playstore_app_data['PRICE']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'KDE plot of price of apps in playstore')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1584x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "playstore_price_kde = stats.gaussian_kde(playstore_app_data['PRICE'])\n",
    "playstore_prices = np.linspace(0, 10, 1000)\n",
    "plt.figure(figsize=(22,8))\n",
    "plt.plot(playstore_prices, playstore_price_kde(playstore_prices), label=\"playstore price\")\n",
    "plt.xlabel(\"Price\")\n",
    "plt.ylabel(\"Probability\")\n",
    "plt.title(\"KDE plot of price of apps in playstore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see from the graph that the price of app on Google Playstore is mainly on range of \\$0 to around \\$6 and the curve narrows down after \\$6 meaning that only few apps have price greater than \\$6.\n",
    "\n",
    "Now lets read another dataset from Apple appstore and perform similar analysis for the apps on Apple appstore."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average price of app in AppStore is  1.726217868556343\n"
     ]
    },
    {
     "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>SN</th>\n",
       "      <th>ID</th>\n",
       "      <th>APP</th>\n",
       "      <th>SIZE_BYTES</th>\n",
       "      <th>CURRENCY</th>\n",
       "      <th>PRICE</th>\n",
       "      <th>RATING_COUNT_TOT</th>\n",
       "      <th>RATING_COUNT_VER</th>\n",
       "      <th>USER_RATING</th>\n",
       "      <th>USER_RATING_VER</th>\n",
       "      <th>VER</th>\n",
       "      <th>CONT_RATING</th>\n",
       "      <th>PRIME_GENRE</th>\n",
       "      <th>SUP_DEVICES.NUM</th>\n",
       "      <th>IPADSC_URLS.NUM</th>\n",
       "      <th>LANG.NUM</th>\n",
       "      <th>VPP_LIC</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>281656475</td>\n",
       "      <td>PAC-MAN Premium</td>\n",
       "      <td>100788224</td>\n",
       "      <td>USD</td>\n",
       "      <td>3.99</td>\n",
       "      <td>21292</td>\n",
       "      <td>26</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.5</td>\n",
       "      <td>6.3.5</td>\n",
       "      <td>4+</td>\n",
       "      <td>Games</td>\n",
       "      <td>38</td>\n",
       "      <td>5</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>281796108</td>\n",
       "      <td>Evernote - stay organized</td>\n",
       "      <td>158578688</td>\n",
       "      <td>USD</td>\n",
       "      <td>0.00</td>\n",
       "      <td>161065</td>\n",
       "      <td>26</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>8.2.2</td>\n",
       "      <td>4+</td>\n",
       "      <td>Productivity</td>\n",
       "      <td>37</td>\n",
       "      <td>5</td>\n",
       "      <td>23</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   SN         ID                        APP  SIZE_BYTES CURRENCY  PRICE  \\\n",
       "0   1  281656475            PAC-MAN Premium   100788224      USD   3.99   \n",
       "1   2  281796108  Evernote - stay organized   158578688      USD   0.00   \n",
       "\n",
       "   RATING_COUNT_TOT  RATING_COUNT_VER  USER_RATING  USER_RATING_VER    VER  \\\n",
       "0             21292                26          4.0              4.5  6.3.5   \n",
       "1            161065                26          4.0              3.5  8.2.2   \n",
       "\n",
       "  CONT_RATING   PRIME_GENRE  SUP_DEVICES.NUM  IPADSC_URLS.NUM  LANG.NUM  \\\n",
       "0          4+         Games               38                5        10   \n",
       "1          4+  Productivity               37                5        23   \n",
       "\n",
       "   VPP_LIC  \n",
       "0        1  \n",
       "1        1  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "apple_appstore_data = pd.read_csv(\"app-store-apple-data-set-10k-apps/AppleStore.csv\")\n",
    "apple_appstore_data.columns = list(map(str.upper, ['SN', 'ID', 'APP', 'size_bytes', 'currency', 'price', 'rating_count_tot', 'rating_count_ver', 'user_rating', 'user_rating_ver', 'ver', 'cont_rating', 'prime_genre', 'sup_devices.num', 'ipadSc_urls.num', 'lang.num', 'vpp_lic']))\n",
    "print (\"Average price of app in AppStore is \", np.average(apple_appstore_data['PRICE']))\n",
    "apple_appstore_data.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lets plot KDE for both price of apps on Google playstore and Apple appstore to compare the price distribution on the platforms."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f60bb3eda20>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1584x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "applestore_price_kde = stats.gaussian_kde(apple_appstore_data['PRICE'])\n",
    "prices = np.linspace(0, 7, 1000)\n",
    "\n",
    "plt.figure(figsize=(22,8))\n",
    "plt.plot(prices, playstore_price_kde(prices), label=\"play store price\")\n",
    "plt.plot(prices, applestore_price_kde(prices), label=\"apple store price\")\n",
    "\n",
    "mean_applestore_price = np.mean(apple_appstore_data['PRICE'])\n",
    "prob_mean_apple = applestore_price_kde(mean_applestore_price)[0]\n",
    "\n",
    "mean_playstore_price = np.mean(playstore_app_data['PRICE'])\n",
    "prob_mean_google = playstore_price_kde(mean_playstore_price)\n",
    "\n",
    "plt.vlines(x=mean_applestore_price, ymin=0, ymax=prob_mean_apple, color='r', label=\"Mean App Store price\")\n",
    "plt.vlines(x=mean_playstore_price, ymin=0, ymax=prob_mean_google, color='g', label=\"Mean Google playstore price\")\n",
    "\n",
    "plt.title(\"Price distribution of apps on Google Playstore and Apple Appstore\")\n",
    "plt.xlabel(\"Price\")\n",
    "plt.ylabel(\"Probability\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the graph, we can see the difference in price distribution of the apps on Google Playstore and Apple Appstore. The mean price of apps on Apple Appstore seems to be higher than mean price of apps on Google Playstore. Is the observed price difference on these platforms significantly different than one other? To answer this question, we can use hypothesis testing."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "## Hypothesis Testing\n",
    "### Price difference between Google Playstore and Apple Appstore\n",
    "\n",
    "##### Null Hypothesis ($H_0$)\n",
    "    The mean values across Google PlayStore and Apple AppStore are the same and any difference is due to solely sampling\n",
    "\n",
    "##### Alternative Hypothesis ($H_A$)\n",
    "    The mean values across Google PlayStore and Apple AppStore are different. The observed difference is not due to chance or sampling.\n",
    "\n",
    "For this, our test statistics will be the difference between mean price of the Play store apps and App store apps. We will assume both Play store apps and App store apps are from same probability distribution and compare out observed test statistics under that distribution to find out the p-value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "all_app_prices = pd.concat([playstore_app_data[['APP', 'PRICE']], apple_appstore_data[['APP', 'PRICE']]])\n",
    "all_app_prices = np.array(all_app_prices['PRICE'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Hist of mean difference between Playstore app price and Appstore app price')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1584x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "price_diffs = []\n",
    "\n",
    "for index in range(5000):\n",
    "    np.random.shuffle(all_app_prices)\n",
    "    google_playstore_shuffled = np.random.choice(all_app_prices, playstore_app_data.shape[0])\n",
    "    apple_appstore_shuffled = np.random.choice(all_app_prices, apple_appstore_data.shape[0])\n",
    "    price_diffs.append(np.average(google_playstore_shuffled) - np.average(apple_appstore_shuffled))\n",
    "\n",
    "observed_test_stat = np.mean(playstore_app_data['PRICE']) - np.mean(apple_appstore_data['PRICE'])\n",
    "plt.figure(figsize=(22,8))\n",
    "plt.hist(price_diffs, label='Price distribution')\n",
    "plt.scatter(observed_test_stat, 0, color='r', label=\"Observed price difference\")\n",
    "plt.xlabel(\"mean difference\")\n",
    "plt.ylabel(\"counts\")\n",
    "plt.legend()\n",
    "plt.title(\"Hist of mean difference between Playstore app price and Appstore app price\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### `p-value` "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.00019999999999997797"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "p_value_price_diffs = sum(np.array(price_diffs) > observed_test_stat) / len(price_diffs)\n",
    "1 - p_value_price_diffs "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see, the `p-value` we obtained from the hypothesis testing is low so we reject the null hypothesis. It means that the observed price difference in Apple Appstore and Google Playstore is significant and the difference is not due to sampling or by chance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Quality of Apps on Playstore"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Free', 'Paid'], dtype=object)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "playstore_app_data_dropna.TYPE.unique()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Google Play store has two types of apps: `Paid` and `Free`. Let's look at the average ratings of the `Paid` and `Free` apps. \n",
    "\n",
    "We will use the rating on the apps to measure the quality of an app. We want to know if free apps have different quality or rating as compared to paid apps. So, lets separate out the `Free` and `Paid` apps and plot KDE to get an overview of how `RATING`s are distributed across apps on the Playstore."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7146, 15)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "free_playstore_apps = playstore_app_data_dropna[playstore_app_data_dropna['TYPE'] == 'Free']\n",
    "free_playstore_apps.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(577, 15)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "paid_playstore_apps = playstore_app_data_dropna[playstore_app_data_dropna['TYPE'] == 'Paid']\n",
    "paid_playstore_apps.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f60bb6a9e48>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "all_app_ratings = np.array(playstore_app_data_dropna['RATING'])\n",
    "paid_app_rating_kde = stats.gaussian_kde(paid_playstore_apps['RATING'])\n",
    "free_app_rating_kde = stats.gaussian_kde(free_playstore_apps['RATING'])\n",
    "ratings = np.linspace(0, 7, 1000)\n",
    "\n",
    "plt.figure(figsize=(16,5))\n",
    "plt.plot(ratings, paid_app_rating_kde(ratings), label=\"Paid apps rating\")\n",
    "plt.plot(ratings, free_app_rating_kde(ratings), label=\"Free apps rating\")\n",
    "\n",
    "mean_free_app_rating = np.mean(free_playstore_apps['RATING'])\n",
    "mean_paid_app_rating = np.mean(paid_playstore_apps['RATING'])\n",
    "\n",
    "prob_mean_free_app_rating = free_app_rating_kde(mean_free_app_rating)[0]\n",
    "prob_mean_paid_app_rating = paid_app_rating_kde(mean_paid_app_rating)[0]\n",
    "\n",
    "plt.vlines(x=mean_free_app_rating, ymin=0, ymax=prob_mean_free_app_rating, color='g', label=\"Mean rating for free app\")\n",
    "plt.vlines(x=mean_paid_app_rating, ymin=0, ymax=prob_mean_paid_app_rating, color='b', label=\"Mean rating for free app\")\n",
    "\n",
    "plt.title(\"Rating distribution of free and paid apps\")\n",
    "plt.xlabel(\"Ratings\")\n",
    "plt.ylabel(\"Probability\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see from above graph the mean rating of paid apps are slightly more than rating of free apps. But we can not infer from the graph that the difference is significant or not. So lets do another hypothesis testing to see if there is significant difference in the average rating of free and paid apps."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Is there significant difference between quality of free apps and paid apps?\n",
    "Lets formulate our hypothesis to answer this question.\n",
    "\n",
    "##### Null Hypothesis ($H_0$)\n",
    "    The mean rating across free apps and paid apps are the same and any difference is solely due to sampling\n",
    "\n",
    "##### Alternative Hypothesis ($H_A$)\n",
    "    The mean rating across free apps and paid apps are different. The observed difference is not due to chance or sampling\n",
    "    \n",
    "For this, our test statistics will be the difference between mean rating of the free and paid apps. We will assume the paid apps and free apps are from same probability distribution and compare out observed test statistics under that distribution to find out the p-value."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.08529734611744821"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(paid_playstore_apps['RATING']) - np.mean(free_playstore_apps['RATING'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f60c6748278>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "rating_diffs = []\n",
    "\n",
    "for index in range(5000):\n",
    "    np.random.shuffle(all_app_ratings)\n",
    "    free_apps_choice = np.random.choice(all_app_ratings, free_playstore_apps.shape[0])\n",
    "    paid_apps_choice = np.random.choice(all_app_ratings, paid_playstore_apps.shape[0])\n",
    "    rating_diffs.append(np.average(paid_apps_choice) - np.average(free_apps_choice))\n",
    "\n",
    "observed_test_stat = np.mean(paid_playstore_apps['RATING']) - np.mean(free_playstore_apps['RATING'])\n",
    "plt.figure(figsize=(18,8))\n",
    "plt.hist(rating_diffs, label=\"Mean difference of ratings\")\n",
    "plt.scatter(observed_test_stat, 0, color='r', label=\"Observed test statistics\")\n",
    "plt.xlabel(\"mean difference\")\n",
    "plt.ylabel(\"counts\")\n",
    "\n",
    "plt.title(\"Hist of mean difference between AVG\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "p-value = 0.0\n"
     ]
    }
   ],
   "source": [
    "p_value_rating_diffs = sum(np.array(rating_diffs) > observed_test_stat) / len(rating_diffs)\n",
    "print (\"p-value =\", p_value_rating_diffs)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The `p-value` is low so we reject the null hypothesis. It means that free app ratings are significantly different than paid app ratings and the difference in ratings is not due to sampling or by chance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Category Classification\n",
    "Every app in our Play store dataset has a category and it defines the type of app. We can use other features in our dataset to classify the apps into its category. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "33"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(playstore_app_data_dropna.CATEGORY.unique())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So we have 33 unique categories in our dataset. Let's build a `LogisticRegression` classifier to help predict the category of app based on other features. We will first normalize our data similar to the normalization scheme we had form predicting `RATING` and then use the normalized dataset to train our `LogisticRegression` classifier to predict the category of our app. We will use grid search to tune our hyperparameters and will make use of 5 fold cross validation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2504854368932039"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "category_encoder = preprocessing.LabelEncoder()\n",
    "category_encoder.fit(playstore_app_data_dropna['CATEGORY'])\n",
    "playstore_app_data_dropna['CATEGORY_ENCODED'] = category_encoder.transform(playstore_app_data_dropna['CATEGORY'])\n",
    "\n",
    "X_data = playstore_app_data_dropna[['REVIEWS', 'PRICE', 'SIZE_NUM', 'INSTALLS_NUM', 'RATING']]\n",
    "y_data = playstore_app_data_dropna['CATEGORY_ENCODED']\n",
    "\n",
    "X_data = normalize_data_for_rating(X_data)\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(X_data, y_data, test_size=0.2, random_state=2)\n",
    "\n",
    "parameters = dict(solver=['lbfgs', 'sag', 'saga'])\n",
    "\n",
    "logistic_model = LogisticRegression(random_state=0, multi_class='multinomial', penalty='l2')\n",
    "lgstc_classifier = GridSearchCV(logistic_model, parameters, cv=5)\n",
    "lgstc_classifier.fit(X_train, y_train)\n",
    "\n",
    "lgstc_classifier.score(X_test, y_test)\n",
    "\n",
    "from sklearn.metrics import accuracy_score\n",
    "accuracy_score(y_test, lgstc_classifier.predict(X_test))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.2504854368932039"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "parameters = dict(n_estimators=[10, 30, 90], criterion=['gini', 'entropy'], max_depth=[1, 3, 9], min_samples_split=[100, 300, 1000])\n",
    "random_forest_model = RandomForestClassifier()\n",
    "rf_classifier = GridSearchCV(random_forest_model, parameters, cv=5)\n",
    "rf_classifier.fit(X_train, y_train)\n",
    "\n",
    "accuracy_score(y_test, rf_classifier.predict(X_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The accuracy of our `LogisticRegressor` classifier and `RandomForestClassifier` is around 25% and we can use the classifier to predict the category of a given app. The observed accuracy of 25% seems to be low but when we take consideration of number of categories we have(33), the accuracy seems to be justifiable.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion\n",
    "In this project, I analyzed the Google Play store apps and used hypothesis testing to examine the significance of the observed in differences in our test statistics.\n",
    "\n",
    "First we trained a linear model to predict the ratings based on other numerical features. Then used that model to impute missing ratings on our dataset. We showed that the price of apps on Google Play store was significantly different than the price of apps on Apple App store. We also found the quality of free apps on Google Play store was significantly different than the quality of paid apps. Finally we built a model to classify category of an app with reasonable accuracy using various features about the app available on the Google Play store.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## References\n",
    "    1. Google Play Store: Number of apps 2018. (n.d.). Retrieved May 4, 2019, from https://www.statista.com/statistics/266210/number-of-available-applications-in-the-google-play-store/ \n",
    "    2. List of most-downloaded Google Play applications. (2019, April 22). Retrieved May 4, 2019, from https://en.wikipedia.org/wiki/List_of_most-downloaded_Google_Play_applications#Free_applications_with_over_five_billion_downloads"
   ]
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}