FinalProject_group095.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# COGS 108 - Final Project "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Names\n",
    "\n",
    "- Sierra Stump\n",
    "- Yisha Zhang\n",
    "- Katie Chen\n",
    "- Austin Labador\n",
    "- Hao Luo"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Overview"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this report, we explore the possible relations between tags (a tool/feature supported by Youtube) and the popularity (measured by number of views) of Youtube videos, in order to find out some possible ways to increase video views by using proper tags. More specifically, we trained two linear regressions models to prove that there is a positive correlation between number of tags and number of views, as well as number of TOP tags (more details below) and number of views. On the other hand, we performed a sentiment analysis to analyze the connotation of the tag words with the highest count. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Research Question"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "How do tags of Youtube videos affect their number of views? \n",
    "- Will the number of tags affect the number of views? \n",
    "- Will the existence of certain popular tags affect the number of views? \n",
    "- Will the connotation of tags affect the number of views?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Background and Prior Work"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Just like any other content distribution network such as Facebook, Youtube has its own recommendation system which selects the \"best\" content to feed to users. Youtubers from all around the world has been trying to figure out how the system works in order to make videos that is liked by both the users and the system (which will eventually attract more users). Some common techniques to increase view are using visually-appealing thumbnails and having click-bait titles. However, these tools that appeal directly to users' eyes might not be able to fool the recommender system, since everything becomes data in the system.\n",
    "\n",
    "Nevertheless, there is one useful tool that a lot of Youtubers often ignore, which is the tags under each video. These tags are editable by the author when uploading the video, and in fact serve as a better source of information (at least to the computer) than things like title and descriptions.\n",
    "\n",
    "We all know that when users search for videos with keywords, they usually search for single words or phrases instead of complete sentences. However, titles and descriptions usually consist of meaningful sentences, which means it's very unlikely that multiple words with similar meanings would appear in the same title or description. For example, people would use either \"Why is smoking harmful\" or \"Why are cigarettes bad for you\" as a title, but no one would use \"Why is/are smoking/cigarettes harmful/bad\". This cause potential audience loss because there are going to be people who search for \"smoking\" and others searching for \"cigarettes\".\n",
    "\n",
    "This wouldn't be a problem for tags because tags are very similar to keywords. You can have multiple tags that has similar meanings for the same video, which will help avoid the problem mentioned above.\n",
    "\n",
    "The potential advantage of using tags is far more than what we mention here, and there are tons of non-acedemic posts online that teach people how to use tags to increase views. \n",
    "\n",
    "Non-acedemic sources:\n",
    "- 1) https://whyvideoisgreat.com/optimize-youtube-tags\n",
    "- 2) https://longtailpro.com/best-tags-for-youtube/\n",
    "\n",
    "In the acedemic field, there are also lots of papers trying to utilize the information provided by tags. For example, the paper \"Enriching and localizing semantic tags in internet videos\" [1] by Lamberto Ballan and others talks about how tags are used to retrieve multimedia content and ease browsing and searching.\n",
    "\n",
    "Another paper titled \"Finding meaning on YouTube: Tag recommendation and category discovery\" [2] by George Toderici and others presents a way to automatically generate tags for Youtube videos.\n",
    "\n",
    "References:\n",
    "- 1) https://dl.acm.org/citation.cfm?id=2072060\n",
    "- 2) https://ieeexplore.ieee.org/abstract/document/4438859\n",
    "\n",
    "Therefore, in this report, we will be exploring the relation between the tags of Youtube videos and their number of views (as a measurement of the popularity of the video)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Hypothesis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Based on our background research:\n",
    "\n",
    "We hypothesize that videos that use more tags will have more views than those with less tags, because more tags means more information for the system to analyze and thus means higher chance to be recommended to users.\n",
    "\n",
    "Similarly, we predict that videos with certain popular tags will have more views than those with none. For example, after the release of a new iPhone, the tag \"iPhone\" might become very popular because of the large number of videos reporting it. At the same time, people are going to be interested in the new iPhone because of its popularity, which might indirectly results in the videos with the tag \"iPhone\" having more views than those that don't.\n",
    "\n",
    "Last but not least, we hypothesize that videos with more positive tag words will have more views than those with negative tag words, because positive content are generally more appealing to people than negative ones."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Dataset(s)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The dataset we use contains daily information concerning top trending videos on YouTube:\n",
    "\n",
    "- Dataset Name: Trending YouTube Video Statistics\n",
    "- Link to the dataset: https://www.kaggle.com/datasnaek/youtube-new\n",
    "- Number of observations: 40949\n",
    "\n",
    "This dataset includes data from several countries, but we are only focusing on the data collected from the United States. Each data row contains the following information: Video ID, Trending Date, Title, Channel Title, Category ID, Publish Time, Tags, Views, Likes, Dislikes, Comment Count, Thumbnail Link, Comments Disabled, Ratings Disabled, Video Error or Removed, and Description."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Setup\n",
    "## Import libraries & tools"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to /home/sstump/nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n",
      "[nltk_data] Downloading package vader_lexicon to\n",
      "[nltk_data]     /home/sstump/nltk_data...\n",
      "[nltk_data]   Package vader_lexicon is already up-to-date!\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "pd.set_option('display.max_rows', 200) # always display 200 rows\n",
    "import matplotlib.pyplot as plt\n",
    "plt.rcParams['figure.figsize'] = (17,7) # increase figure size\n",
    "import seaborn as sns\n",
    "sns.set(style ='white', font_scale=2) #set style\n",
    "from scipy.stats import uniform, norm, bernoulli, poisson\n",
    "import re\n",
    "from nltk.tokenize import WhitespaceTokenizer\n",
    "from nltk.corpus import stopwords\n",
    "import nltk\n",
    "nltk.download('stopwords')\n",
    "from nltk.stem import PorterStemmer\n",
    "from nltk.probability import FreqDist\n",
    "from nltk.sentiment.vader import SentimentIntensityAnalyzer\n",
    "nltk.download('vader_lexicon')\n",
    "import string\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading Data"
   ]
  },
  {
   "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>video_id</th>\n",
       "      <th>trending_date</th>\n",
       "      <th>title</th>\n",
       "      <th>channel_title</th>\n",
       "      <th>category_id</th>\n",
       "      <th>publish_time</th>\n",
       "      <th>tags</th>\n",
       "      <th>views</th>\n",
       "      <th>likes</th>\n",
       "      <th>dislikes</th>\n",
       "      <th>comment_count</th>\n",
       "      <th>thumbnail_link</th>\n",
       "      <th>comments_disabled</th>\n",
       "      <th>ratings_disabled</th>\n",
       "      <th>video_error_or_removed</th>\n",
       "      <th>description</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2kyS6SvSYSE</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>WE WANT TO TALK ABOUT OUR MARRIAGE</td>\n",
       "      <td>CaseyNeistat</td>\n",
       "      <td>22</td>\n",
       "      <td>2017-11-13T17:13:01.000Z</td>\n",
       "      <td>SHANtell martin</td>\n",
       "      <td>748374</td>\n",
       "      <td>57527</td>\n",
       "      <td>2966</td>\n",
       "      <td>15954</td>\n",
       "      <td>https://i.ytimg.com/vi/2kyS6SvSYSE/default.jpg</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>SHANTELL'S CHANNEL - https://www.youtube.com/s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1ZAPwfrtAFY</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>The Trump Presidency: Last Week Tonight with J...</td>\n",
       "      <td>LastWeekTonight</td>\n",
       "      <td>24</td>\n",
       "      <td>2017-11-13T07:30:00.000Z</td>\n",
       "      <td>last week tonight trump presidency|\"last week ...</td>\n",
       "      <td>2418783</td>\n",
       "      <td>97185</td>\n",
       "      <td>6146</td>\n",
       "      <td>12703</td>\n",
       "      <td>https://i.ytimg.com/vi/1ZAPwfrtAFY/default.jpg</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>One year after the presidential election, John...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5qpjK5DgCt4</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>Racist Superman | Rudy Mancuso, King Bach &amp; Le...</td>\n",
       "      <td>Rudy Mancuso</td>\n",
       "      <td>23</td>\n",
       "      <td>2017-11-12T19:05:24.000Z</td>\n",
       "      <td>racist superman|\"rudy\"|\"mancuso\"|\"king\"|\"bach\"...</td>\n",
       "      <td>3191434</td>\n",
       "      <td>146033</td>\n",
       "      <td>5339</td>\n",
       "      <td>8181</td>\n",
       "      <td>https://i.ytimg.com/vi/5qpjK5DgCt4/default.jpg</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>WATCH MY PREVIOUS VIDEO ▶ \\n\\nSUBSCRIBE ► http...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>puqaWrEC7tY</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>Nickelback Lyrics: Real or Fake?</td>\n",
       "      <td>Good Mythical Morning</td>\n",
       "      <td>24</td>\n",
       "      <td>2017-11-13T11:00:04.000Z</td>\n",
       "      <td>rhett and link|\"gmm\"|\"good mythical morning\"|\"...</td>\n",
       "      <td>343168</td>\n",
       "      <td>10172</td>\n",
       "      <td>666</td>\n",
       "      <td>2146</td>\n",
       "      <td>https://i.ytimg.com/vi/puqaWrEC7tY/default.jpg</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>Today we find out if Link is a Nickelback amat...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>d380meD0W0M</td>\n",
       "      <td>17.14.11</td>\n",
       "      <td>I Dare You: GOING BALD!?</td>\n",
       "      <td>nigahiga</td>\n",
       "      <td>24</td>\n",
       "      <td>2017-11-12T18:01:41.000Z</td>\n",
       "      <td>ryan|\"higa\"|\"higatv\"|\"nigahiga\"|\"i dare you\"|\"...</td>\n",
       "      <td>2095731</td>\n",
       "      <td>132235</td>\n",
       "      <td>1989</td>\n",
       "      <td>17518</td>\n",
       "      <td>https://i.ytimg.com/vi/d380meD0W0M/default.jpg</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>I know it's been a while since we did this sho...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      video_id trending_date  \\\n",
       "0  2kyS6SvSYSE      17.14.11   \n",
       "1  1ZAPwfrtAFY      17.14.11   \n",
       "2  5qpjK5DgCt4      17.14.11   \n",
       "3  puqaWrEC7tY      17.14.11   \n",
       "4  d380meD0W0M      17.14.11   \n",
       "\n",
       "                                               title          channel_title  \\\n",
       "0                 WE WANT TO TALK ABOUT OUR MARRIAGE           CaseyNeistat   \n",
       "1  The Trump Presidency: Last Week Tonight with J...        LastWeekTonight   \n",
       "2  Racist Superman | Rudy Mancuso, King Bach & Le...           Rudy Mancuso   \n",
       "3                   Nickelback Lyrics: Real or Fake?  Good Mythical Morning   \n",
       "4                           I Dare You: GOING BALD!?               nigahiga   \n",
       "\n",
       "   category_id              publish_time  \\\n",
       "0           22  2017-11-13T17:13:01.000Z   \n",
       "1           24  2017-11-13T07:30:00.000Z   \n",
       "2           23  2017-11-12T19:05:24.000Z   \n",
       "3           24  2017-11-13T11:00:04.000Z   \n",
       "4           24  2017-11-12T18:01:41.000Z   \n",
       "\n",
       "                                                tags    views   likes  \\\n",
       "0                                    SHANtell martin   748374   57527   \n",
       "1  last week tonight trump presidency|\"last week ...  2418783   97185   \n",
       "2  racist superman|\"rudy\"|\"mancuso\"|\"king\"|\"bach\"...  3191434  146033   \n",
       "3  rhett and link|\"gmm\"|\"good mythical morning\"|\"...   343168   10172   \n",
       "4  ryan|\"higa\"|\"higatv\"|\"nigahiga\"|\"i dare you\"|\"...  2095731  132235   \n",
       "\n",
       "   dislikes  comment_count                                  thumbnail_link  \\\n",
       "0      2966          15954  https://i.ytimg.com/vi/2kyS6SvSYSE/default.jpg   \n",
       "1      6146          12703  https://i.ytimg.com/vi/1ZAPwfrtAFY/default.jpg   \n",
       "2      5339           8181  https://i.ytimg.com/vi/5qpjK5DgCt4/default.jpg   \n",
       "3       666           2146  https://i.ytimg.com/vi/puqaWrEC7tY/default.jpg   \n",
       "4      1989          17518  https://i.ytimg.com/vi/d380meD0W0M/default.jpg   \n",
       "\n",
       "   comments_disabled  ratings_disabled  video_error_or_removed  \\\n",
       "0              False             False                   False   \n",
       "1              False             False                   False   \n",
       "2              False             False                   False   \n",
       "3              False             False                   False   \n",
       "4              False             False                   False   \n",
       "\n",
       "                                         description  \n",
       "0  SHANTELL'S CHANNEL - https://www.youtube.com/s...  \n",
       "1  One year after the presidential election, John...  \n",
       "2  WATCH MY PREVIOUS VIDEO ▶ \\n\\nSUBSCRIBE ► http...  \n",
       "3  Today we find out if Link is a Nickelback amat...  \n",
       "4  I know it's been a while since we did this sho...  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"USvideos.csv\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(40949, 16)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Size of data set: used to verify that all of the data was read in correctly\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Date Wraggling and Cleaning\n",
    "## 1. Date reformatting\n",
    "Changing date format of both \"trending_date\" and \"publish_time\" to yyyy-mm-dd so that it's easier to read and order"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['trending_date'] = pd.to_datetime(df['trending_date'], format='%y.%d.%m').dt.date\n",
    "df['publish_time'] = pd.to_datetime(df['publish_time'], format='%Y-%m-%dT%H:%M:%S.%fZ').dt.date"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Drop irrelevant columns\n",
    "Since we are only curious about the relation between tags and views, a lot of information in the dataframe are actually irrelevant. Therefore, we decide to remove them from the data frame so that we don't need to worry about tham again when handling missing values.\n",
    "\n",
    "However, we are keeping the likes, dislikes, and comment_count columns because they can also be measurement of \"popularity\" of a video, so they are potentially useful if the data of views is undermined."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.drop(labels = ['video_id', 'thumbnail_link', 'channel_title', 'category_id', \n",
    "                  'description', 'comments_disabled', 'ratings_disabled'], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Drop error or removed videos\n",
    "Delete videos that are removed or have errors because these rows might contain invalid/wrong data, which may falsify our analysis later."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df[df['video_error_or_removed'] == False]\n",
    "df.drop('video_error_or_removed',axis=1,inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Missing values\n",
    "After an initial observation it seems like there is no NaN values in the tags column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n"
     ]
    }
   ],
   "source": [
    "n_nan_tags =  sum(df['tags'].isnull())\n",
    "print(n_nan_tags)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "However, if we look more closely, there are actually lots of videos that have no tags. These videos have \"none\" as the value for the tags column. Since it's very difficult for us to verify whether these videos actually don't have tags or the author of this dataset fail to retrieve this information, it would be better to not use this part of the data. Therefore, we decide to remove it from the data frame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>trending_date</th>\n",
       "      <th>title</th>\n",
       "      <th>publish_time</th>\n",
       "      <th>tags</th>\n",
       "      <th>views</th>\n",
       "      <th>likes</th>\n",
       "      <th>dislikes</th>\n",
       "      <th>comment_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>Dennis Smith Jr. and LeBron James go back and ...</td>\n",
       "      <td>2017-11-13</td>\n",
       "      <td>[none]</td>\n",
       "      <td>945</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>Edna's registered owner thought she was dead f...</td>\n",
       "      <td>2017-11-10</td>\n",
       "      <td>[none]</td>\n",
       "      <td>284666</td>\n",
       "      <td>16396</td>\n",
       "      <td>81</td>\n",
       "      <td>949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>President Bill Clinton On Dictators, Democracy...</td>\n",
       "      <td>2017-11-09</td>\n",
       "      <td>[none]</td>\n",
       "      <td>366180</td>\n",
       "      <td>4364</td>\n",
       "      <td>4448</td>\n",
       "      <td>1997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>136</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>Alicia Keys - When You Were Gone</td>\n",
       "      <td>2017-11-09</td>\n",
       "      <td>[none]</td>\n",
       "      <td>95944</td>\n",
       "      <td>1354</td>\n",
       "      <td>181</td>\n",
       "      <td>117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>175</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>Sphaera - demonstrating interaction</td>\n",
       "      <td>2017-11-04</td>\n",
       "      <td>[none]</td>\n",
       "      <td>1827</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    trending_date                                              title  \\\n",
       "42     2017-11-14  Dennis Smith Jr. and LeBron James go back and ...   \n",
       "97     2017-11-14  Edna's registered owner thought she was dead f...   \n",
       "133    2017-11-14  President Bill Clinton On Dictators, Democracy...   \n",
       "136    2017-11-14                   Alicia Keys - When You Were Gone   \n",
       "175    2017-11-14                Sphaera - demonstrating interaction   \n",
       "\n",
       "    publish_time    tags   views  likes  dislikes  comment_count  \n",
       "42    2017-11-13  [none]     945      7         5              8  \n",
       "97    2017-11-10  [none]  284666  16396        81            949  \n",
       "133   2017-11-09  [none]  366180   4364      4448           1997  \n",
       "136   2017-11-09  [none]   95944   1354       181            117  \n",
       "175   2017-11-04  [none]    1827      3         0              2  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df['tags'] == \"[none]\"].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1521"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df[df['tags'] == \"[none]\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Just in case we want to use it for comparison later, we will also make a copy of the \n",
    "# data set with the \"none\" values first, and save it in 'df_withNoTags'\n",
    "df_withNoTags = df.copy()\n",
    "df = df[df['tags'] != \"[none]\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Tags and Title format clean up"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Convert all string data to lowercase and strip all the extra/useless spaces so that they are easier to handle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# function that strip the white space of the string\n",
    "def strip(name) :\n",
    "    return str(name).strip()\n",
    "\n",
    "# function that turn string to lower case\n",
    "def lower_case(name):\n",
    "    return str(name).lower()\n",
    "\n",
    "# strip the leading and ending white space in tags column\n",
    "df['tags'] = df['tags'].apply(strip)\n",
    "df['title'] = df['title'].apply(strip)\n",
    "\n",
    "# change the tags column to all lower case lettering\n",
    "df[\"tags\"] = df[\"tags\"].apply(lower_case)\n",
    "df[\"title\"] = df[\"title\"].apply(lower_case)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5. Dropping rows where number of views is too large\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df[df['views'] <= 20000000]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to make the graph easier to visualize, we dropped the videos that have over 20 million views. We consider these videos outliers because the majority of the other videos have views around 7 million or less. When examining the videos with 20 million views, we observed that most of them are music videos. It makes sense why music videos would have a significantly larger amount of views when compared to videos posted by ordinary YouTubers. We decided to drop this data because it doesn’t significantly change the shape of the graph, therefore the resulting analysis should not change. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Take a look at what the data looks like after cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>trending_date</th>\n",
       "      <th>title</th>\n",
       "      <th>publish_time</th>\n",
       "      <th>tags</th>\n",
       "      <th>views</th>\n",
       "      <th>likes</th>\n",
       "      <th>dislikes</th>\n",
       "      <th>comment_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>we want to talk about our marriage</td>\n",
       "      <td>2017-11-13</td>\n",
       "      <td>shantell martin</td>\n",
       "      <td>748374</td>\n",
       "      <td>57527</td>\n",
       "      <td>2966</td>\n",
       "      <td>15954</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>the trump presidency: last week tonight with j...</td>\n",
       "      <td>2017-11-13</td>\n",
       "      <td>last week tonight trump presidency|\"last week ...</td>\n",
       "      <td>2418783</td>\n",
       "      <td>97185</td>\n",
       "      <td>6146</td>\n",
       "      <td>12703</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>racist superman | rudy mancuso, king bach &amp; le...</td>\n",
       "      <td>2017-11-12</td>\n",
       "      <td>racist superman|\"rudy\"|\"mancuso\"|\"king\"|\"bach\"...</td>\n",
       "      <td>3191434</td>\n",
       "      <td>146033</td>\n",
       "      <td>5339</td>\n",
       "      <td>8181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>nickelback lyrics: real or fake?</td>\n",
       "      <td>2017-11-13</td>\n",
       "      <td>rhett and link|\"gmm\"|\"good mythical morning\"|\"...</td>\n",
       "      <td>343168</td>\n",
       "      <td>10172</td>\n",
       "      <td>666</td>\n",
       "      <td>2146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>i dare you: going bald!?</td>\n",
       "      <td>2017-11-12</td>\n",
       "      <td>ryan|\"higa\"|\"higatv\"|\"nigahiga\"|\"i dare you\"|\"...</td>\n",
       "      <td>2095731</td>\n",
       "      <td>132235</td>\n",
       "      <td>1989</td>\n",
       "      <td>17518</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  trending_date                                              title  \\\n",
       "0    2017-11-14                 we want to talk about our marriage   \n",
       "1    2017-11-14  the trump presidency: last week tonight with j...   \n",
       "2    2017-11-14  racist superman | rudy mancuso, king bach & le...   \n",
       "3    2017-11-14                   nickelback lyrics: real or fake?   \n",
       "4    2017-11-14                           i dare you: going bald!?   \n",
       "\n",
       "  publish_time                                               tags    views  \\\n",
       "0   2017-11-13                                    shantell martin   748374   \n",
       "1   2017-11-13  last week tonight trump presidency|\"last week ...  2418783   \n",
       "2   2017-11-12  racist superman|\"rudy\"|\"mancuso\"|\"king\"|\"bach\"...  3191434   \n",
       "3   2017-11-13  rhett and link|\"gmm\"|\"good mythical morning\"|\"...   343168   \n",
       "4   2017-11-12  ryan|\"higa\"|\"higatv\"|\"nigahiga\"|\"i dare you\"|\"...  2095731   \n",
       "\n",
       "    likes  dislikes  comment_count  \n",
       "0   57527      2966          15954  \n",
       "1   97185      6146          12703  \n",
       "2  146033      5339           8181  \n",
       "3   10172       666           2146  \n",
       "4  132235      1989          17518  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Visualization\n",
    "## 1. Histograms of numerical data\n",
    "First let's take a look at the distribution of the number of views of all the videos."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA5sAAAJoCAYAAADoGvUTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3X2QlfV99/EP+yQoEBNZARUFBoOCxCCKoI21EQ2TuKaRJBNAEIegeXDGO4WYVB0LWu/kTsGHNrFSNEBBtu3UECUmMD4QjeMmrBVw0S3BVlQkaxUHDaiw7O79h7NbtruYBS5CGl+vfzhc1+97zu/sf++5zrlOt5aWlpYAAABAgUoO9wYAAAD44yM2AQAAKJzYBAAAoHBiEwAAgMKJTQAAAAonNgEAACic2AQAAKBwYhMAAIDCiU0AAAAKJzYBAAAonNgEAACgcGWHewN/6N59991s2LAhlZWVKS0tPdzbAQAA+L1qamrKa6+9ltNOOy3du3fv8pzY/B02bNiQyZMnH+5tAAAAHFb33ntvzjzzzC6vF5u/Q2VlZZL3/rD9+vU7zLsBAAD4/WpoaMjkyZPb2qirxObv0PrR2X79+uWEE044zLsBAAA4PPb3a4VuEAQAAEDhxCYAAACFE5sAAAAUTmwCAABQOLEJAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhxCYAAACFE5sAAAAUTmwCAABQOLEJAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObP4vtruxqdPHAAAAh1vZ4d4AB66ivDRVM+9PkqyY99nDvBsAAID/5somAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhyrq6cMmSJXnqqafy61//Om+88UZ27NiRXr165ZRTTsnnPve5XHLJJenWrVu7mSlTpmTNmjX7fM4/+ZM/yT333NPpud27d+eee+7Jgw8+mJdffjlHHHFERowYkWnTpuUTn/jEPp+zubk51dXVue+++/LCCy+kpKQkQ4cOzaRJk3LxxRd39e0CAABwELocmwsWLMgbb7yRk08+OSNHjkyPHj2ydevW/PKXv0xNTU1WrVqV73//+ykp6Xix9E/+5E9SWVnZ4fhHP/rRTl/r7bffzuWXX55nnnkmH/nIR/Knf/qneeutt1JTU5Mnnngi3/72t3PFFVd0mGtqasrVV1+dRx99ND179sy5556b3bt3p6amJjNnzsy6detyww03dPUtAwAAcIC6HJu33nprhg0bliOPPLLd8U2bNmXatGl55JFHsnz58kyYMKHD7JVXXpmzzz67y5uaN29ennnmmYwePTp33XVXjjrqqCTJ+vXrc/nll+f//b//l7PPPjvDhg1rN7d48eI8+uijGTJkSBYvXpw+ffokSTZv3pzJkydnyZIlGTNmTMaNG9flvQAAALD/uvydzTPPPLNDaCbJySefnEmTJiVJnnzyyYPe0Pbt2/PP//zPKSkpyf/9v/+3LTST5PTTT8+Xv/zltLS0ZP78+e3mmpqacvfddydJZs+e3RaaSTJw4MDMmjUrSXLXXXcd9B4BAAB4f4XcIKis7L0LpBUVFQf9XI899lgaGxszcuTIDBgwoMP5qqqqJMnjjz+exsbGtuNr167Ntm3b0q9fv5x11lkd5saPH5/y8vLU1dXl1VdfPeh9AgAAsG9d/hjtvrz88sv5p3/6pyTJJz/5yU7XPPTQQ3nooYeye/fuHHvssRkzZkzOPPPMTtfW19cnSUaMGNHp+ZNOOikf+tCH8uabb2bz5s05+eSTuzTXo0ePDBkyJPX19amvr0/fvn27/iYBAADYL/sdm/fdd19qa2vT2NiYV199NWvXrk1zc3OuuuqqXHjhhZ3OLFmypN3//+7v/i5nnHFGbr311vTv37/duS1btiRJjjvuuH3uoV+/fnnzzTezZcuWttjsylz//v1TX1/fthYAAIBDY79j8+mnn87y5cv/+wnKynLNNdd0enfYUaNG5bOf/WzOPPPM9OvXL2+88UbWrl2bW2+9NU8//XSmTZuW5cuXt/su6Ntvv50knX4/tFXruZ07d3aY69Gjx37NAQAAULz9/s7mLbfcko0bN2b9+vV58MEHM3Xq1Hz/+9/PF7/4xQ7fhfw//+f/5POf/3wGDhyY7t2757jjjstnPvOZLF++PAMGDMjmzZtTXV3dbqalpeWA3kjr3P/8rU8AAAB+/w74BkHdu3fPkCFD8q1vfSt/8Rd/kX//93/PzTff3KXZ3r17Z+rUqUneuyHQ3lrvPtt6pbIzref2vlPtgc4BAABQvELuRnvppZcmSVavXt3uDrHvZ/DgwUnS4Wro8ccfnyTZunXrPmcbGhrarT2YOQAAAIpXSGz27t07ZWVl2bNnT958880uzWzfvj1Jx6uMw4YNS5LU1dV1Ovfiiy/mzTffTI8ePTJw4MAuz73zzjvZtGlTu7UAAAAcGoXEZm1tbfbs2ZPevXvnwx/+cJdmfvaznyVJTjvttHbH//RP/zTl5eVZu3ZtXn755Q5zK1asaFu39+96jhw5Msccc0waGhpSW1vbYW7lypVpbGzMiBEj/OwJAADAIdal2HzqqafywAMPZPfu3R3O/du//Vuuv/76JMnnP//5lJaWJkl+9atfZc2aNR1u+PPOO+/ke9/7Xh5++OGUlZVlypQp7c4fffTR+eIXv5jm5uZcd9117e4cu379+tx9993p1q1brrrqqnZzpaWlmT59epJk9uzZ2bZtW9u5zZs3Z968eUmSr3zlK115ywAAAByELv30yUsvvZS//Mu/zM0335xhw4alT58+2blzZ15++eU8//zzSZLzzz8/11xzTdtMfX19vvOd76RPnz456aSTcuyxx2b79u2pr6/P9u3bU1FRkVtuuaXtdzL3NmvWrDzzzDNZs2ZNLrzwwpx11ln57W9/m1/+8pdpamrKt771rU4/Cjtt2rTU1tZm9erVueiiizJ27Njs2bMnTz75ZHbt2pUpU6Zk3LhxB/q3AgAAoIu6FJtnnXVWvva1r+Wpp57K5s2bs3bt2rS0tKSysjKf+tSncskll3SIuNGjR+dLX/pSNmzYkBdffDHPPPNMysrKcvzxx+fiiy/OZZddlkGDBnX6ekceeWTuvffe3HPPPfnJT36S1atX54gjjsiYMWNyxRVX5BOf+ESnc6WlpbnzzjuzbNmy/OhHP8oTTzyRkpKSDB8+PJMmTUpVVdV+/nkAAAA4EN1aDvSHLT8gtmzZkgsuuCCPPPJITjjhhMO9nQ6qZt6fJFkx77OHeScAAMAfowNtokJuEAQAAAB7E5sAAAAUTmwCAABQOLEJAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhxCYAAACFE5sAAAAUTmwCAABQOLEJAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhxCYAAACFE5sAAAAUTmwCAABQOLEJAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhxCYAAACFE5sAAAAUTmwCAABQOLEJAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhxCYAAACFE5sAAAAUTmwCAABQOLEJAABA4cQmAAAAhRObAAAAFK6sqwuXLFmSp556Kr/+9a/zxhtvZMeOHenVq1dOOeWUfO5zn8sll1ySbt26dZhrbm5OdXV17rvvvrzwwgspKSnJ0KFDM2nSpFx88cXv+5orVqxIdXV1Nm7cmObm5gwaNCgTJkzIxIkTU1Ky705+/PHHs2jRomzYsCG7du3KgAED8pnPfCbTp09PRUVFV98yAAAAB6jLsblgwYK88cYbOfnkkzNy5Mj06NEjW7duzS9/+cvU1NRk1apV+f73v98uApuamnL11Vfn0UcfTc+ePXPuuedm9+7dqampycyZM7Nu3brccMMNnb7enDlzsmzZshxxxBEZO3ZsysrKUlNTk5tuuik1NTW54447Ulpa2uk+586dm9LS0owePTq9e/dObW1tbr/99vz85z/PokWL0qNHjwP4UwEAANBVXY7NW2+9NcOGDcuRRx7Z7vimTZsybdq0PPLII1m+fHkmTJjQdm7x4sV59NFHM2TIkCxevDh9+vRJkmzevDmTJ0/OkiVLMmbMmIwbN67dc65atSrLli1LZWVlli5dmoEDByZJXn/99UydOjUPPfRQli5dmssvv7zdXF1dXebNm5cePXpk8eLFOf3005MkO3fuzFVXXZXa2trcdtttue6667r+FwIAAGC/dfk7m2eeeWaH0EySk08+OZMmTUqSPPnkk23Hm5qacvfddydJZs+e3RaaSTJw4MDMmjUrSXLXXXd1eM758+cnSWbNmtUWmknSp0+fzJ49O8l7VzCbm5vbzS1YsCAtLS358pe/3BaaSXLUUUflO9/5TkpKSrJs2bK89dZbXX3bAAAAHIBCbhBUVvbeBdK9vw+5du3abNu2Lf369ctZZ53VYWb8+PEpLy9PXV1dXn311bbjDQ0NefbZZ1NeXp7x48d3mBs9enT69u2b1157LevWrWs7vnv37jz++ONJkksuuaTD3IABA/Lxj388jY2Neeyxxw78zQIAAPA7HXRsvvzyy/mnf/qnJMknP/nJtuP19fVJkhEjRnQ616NHjwwZMqTd2iR57rnnkrx3xbR79+6dzrY+595zL7zwQt55550cffTROfHEE993rvU1AAAAODS6/J3NVvfdd19qa2vT2NiYV199NWvXrk1zc3OuuuqqXHjhhW3rtmzZkiQ57rjj9vlc/fv3T319fdva/Znbe+3ej1vPdab1OV955ZV9rgEAAODg7XdsPv3001m+fPl/P0FZWa655ppcccUV7da9/fbbSfK+d35t/Q7ozp0792vuqKOOOqC5zl4PAACA4u33x2hvueWWbNy4MevXr8+DDz6YqVOn5vvf/36++MUvtvvuZUtLS5J0+tub7+f3PQcAAEDxDvg7m927d8+QIUPyrW99K3/xF3+Rf//3f8/NN9/cdr716mPrFcfOtJ5rXdvVudYrk/s719nrAQAAULxC7kZ76aWXJklWr16dxsbGJMnxxx+fJNm6des+5xoaGtqtLWLuN7/5zT7nWs/tPQcAAEDxConN3r17p6ysLHv27Mmbb76ZJBk2bFiSpK6urtOZd955J5s2bWq3du/HmzZtyrvvvtvpbOtznnrqqW3HBg8enO7du2f79u156aWXOp175plnOswBAABQvEJis7a2Nnv27Env3r3z4Q9/OEkycuTIHHPMMWloaEhtbW2HmZUrV6axsTEjRoxI37592473798/w4cPT2NjY1auXNlhbs2aNWloaEhlZWVGjhzZdryioiLnnXdekuSBBx7oMPfyyy9n3bp1KS8vz/nnn3+wbxkAAID30aXYfOqpp/LAAw9k9+7dHc7927/9W66//vokyec///mUlpYmSUpLSzN9+vQkyezZs7Nt27a2mc2bN2fevHlJkq985SsdnvPKK69MksydOzcvvvhi2/Ft27Zlzpw5SZIZM2akpKT99mfMmJFu3brl7rvvbruKmbz3Hc/rrrsuzc3NmTRpUnr37t2Vtw0AAMAB6tJPn7z00kv5y7/8y9x8880ZNmxY+vTpk507d+bll1/O888/nyQ5//zzc80117SbmzZtWmpra7N69epcdNFFGTt2bPbs2ZMnn3wyu3btypQpUzJu3LgOrzd+/PhMnDgx1dXVqaqqyjnnnJOysrLU1NRkx44dGTduXC677LIOcx/72Mcyc+bMzJ07N1/60pcyZsyY9OrVK7W1tdm2bVtOP/30fOMb3ziQvxMAAAD7oUuxedZZZ+VrX/tannrqqWzevDlr165NS0tLKisr86lPfSqXXHJJp9FYWlqaO++8M8uWLcuPfvSjPPHEEykpKcnw4cMzadKkVFVV7fM1Z8+enVGjRuXee+/NmjVr0tzcnMGDB2fChAmZOHFih6uarWbMmJGhQ4dm4cKFqaury65duzJgwIBMmTIl06dPT0VFRRf/NAAAAByobi2tP1BJp7Zs2ZILLrggjzzySE444YTDvZ0OqmbenyRZMe+zh3knAADAH6MDbaJCbhAEAAAAexObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhxCYAAACFE5sAAAAUTmwCAABQOLEJAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhxCYAAACFE5sAAAAUTmwCAABQOLEJAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhxCYAAACFE5sAAAAUTmwCAABQOLEJAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhxCYAAACFE5sAAAAUTmwCAABQOLEJAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFC4sq4samxszFNPPZXHHnssTz/9dLZu3Zrt27fnwx/+cEaOHJnJkyfn7LPP7jD37W9/O8uXL9/n8w4aNCgrV67s9Fxzc3Oqq6tz33335YUXXkhJSUmGDh2aSZMm5eKLL37f/a5YsSLV1dXZuHFjmpubM2jQoEyYMCETJ05MSYm+BgAAONS6FJu1tbW54oorkiSVlZUZPnx4evTokf/4j//IqlWrsmrVqnzta1/LNddc0+n8GWeckZNOOqnD8crKyk7XNzU15eqrr86jjz6anj175txzz83u3btTU1OTmTNnZt26dbnhhhs6nZ0zZ06WLVuWI444ImPHjk1ZWVlqampy0003paamJnfccUdKS0u78rYBAAA4QF2KzW7duuVTn/pUpk6dmjPPPLPduZ/+9KeZNWtW7rzzzpx99tkZM2ZMh/kvfOELufTSS7u8qcWLF+fRRx/NkCFDsnjx4vTp0ydJsnnz5kyePDlLlizJmDFjMm7cuHZzq1atyrJly1JZWZmlS5dm4MCBSZLXX389U6dOzUMPPZSlS5fm8ssv7/JeAAAA2H9d+kzp2LFj87d/+7cdQjNJPv3pT+dzn/tckuSBBx446A01NTXl7rvvTpLMnj27LTSTZODAgZk1a1aS5K677uowO3/+/CTJrFmz2kIzSfr06ZPZs2cnSRYsWJDm5uaD3icAAAD7VsgXGIcNG5YkefXVVw/6udauXZtt27alX79+OeusszqcHz9+fMrLy1NXV9fu9RoaGvLss8+mvLw848eP7zA3evTo9O3bN6+99lrWrVt30PsEAABg37r0MdrfZfPmzUn2/R3MX/3qV9m4cWPefvvtHHPMMRk1alTOPffcTm/WU19fnyQZMWJEp8/Vo0ePDBkyJPX19amvr0/fvn2TJM8991yS5OSTT0737t07nR0xYkReffXV1NfX54wzztiv9wgAAEDXHXRsvvbaa213nL3ooos6XfPjH/+4w7EhQ4bk1ltvzdChQ9sd37JlS5LkuOOO2+dr9u/fP/X19W1r92du77UAAAAcGgf1Mdo9e/bkm9/8Zn77299m7Nix+eQnP9nu/CmnnJIbbrghDz74YNauXZtf/OIXmT9/fk455ZQ8//zzueKKKzp89Pbtt99O8t4VzH058sgjkyQ7d+7cr7mjjjqqwxwAAADFO6grm3/1V3+Vmpqa9O/fP3/zN3/T4fy0adPa/f/II4/Msccem3POOSdTpkzJunXrMn/+/Nx4441ta1paWpK8dwfc/XGgcwAAABTvgK9s/vVf/3X+9V//NZWVlVm0aNE+v6/ZmYqKilx55ZVJkscee6zdudarj61XKjvTeq51bVfnWq9o7j0HAABA8Q4oNr/73e9myZIl+chHPpJFixa1+5mRrho8eHCSjnewPf7445MkW7du3edsQ0NDu7UHMwcAAEDx9js2v/e972XhwoU5+uijs3DhwgwZMuSAXnj79u1JOl5lbP0Zlbq6uk7n3nnnnWzatKnd2r0fb9q0Ke+++26ns63Peeqppx7QngEAAOia/YrNuXPn5p577smHPvShLFy4MKeccsoBv/DPfvazJMlpp53W7vjIkSNzzDHHpKGhIbW1tR3mVq5cmcbGxowYMaLtZ0+S9+40O3z48DQ2NmblypUd5tasWZOGhoZUVlZm5MiRB7xvAAAAfrcux+btt9+eBQsWpHfv3vnhD3/Y7qpiZ+rr67N69eo0NTW1O75nz54sXLgwS5YsSdLxJkKlpaWZPn16kmT27NnZtm1b27nNmzdn3rx5SZKvfOUrHV6z9Xugc+fOzYsvvth2fNu2bZkzZ06SZMaMGZ3+vicAAADF6dLdaB955JH8/d//fZLkxBNPzNKlSztdN3jw4Lbge+WVV/L1r389Rx99dAYOHJi+fftm586d+fWvf53/+q//SklJSWbNmpVPfOITHZ5n2rRpqa2tzerVq3PRRRdl7Nix2bNnT5588sns2rUrU6ZMybhx4zrMjR8/PhMnTkx1dXWqqqpyzjnnpKysLDU1NdmxY0fGjRuXyy67rMt/HAAAAA5Ml2LzzTffbHu8YcOGbNiwodN1o0ePbovNoUOHZurUqamrq8srr7yS5557Lt26dUu/fv1y6aWXZvLkyR0+QtuqtLQ0d955Z5YtW5Yf/ehHeeKJJ1JSUpLhw4dn0qRJqaqq2udeZ8+enVGjRuXee+/NmjVr0tzcnMGDB2fChAmZOHGiq5oAAAC/B91aWn+gkk5t2bIlF1xwQR555JGccMIJh3s7HVTNvD9Jct93L05FeWl2Nzalorz0MO8KAAD4Y3GgTeQy3x+JivLSVM28X2gCAAB/EMQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhxCYAAACFE5sAAAAUTmwCAABQOLEJAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhxCYAAACFE5t/ZHY3NnX6GAAA4Pep7HBvgGJVlJemaub9SZIV8z57mHcDAAB8ULmyCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhxCYAAACFE5sAAAAUTmwCAABQOLEJAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhxCYAAACFE5sAAAAUTmwCAABQOLH5R2x3Y1O7fwEAAH5fxOYfsYry0lTNvD8V5aWHeysAAMAHjNgEAACgcGITAACAwolNAAAACic2AQAAKJzYBAAAoHBi8wNg758+8TMoAADA70PZ4d4Ah17rT6AkyYp5nz3MuwEAAD4IXNkEAACgcF26stnY2Jinnnoqjz32WJ5++uls3bo127dvz4c//OGMHDkykydPztlnn73P+RUrVqS6ujobN25Mc3NzBg0alAkTJmTixIkpKdl37z7++ONZtGhRNmzYkF27dmXAgAH5zGc+k+nTp6eiomKfc+vXr88//MM/5Omnn86OHTvSv3//jBs3Ll/96lfTq1evrrxlAAAADkKXYrO2tjZXXHFFkqSysjLDhw9Pjx498h//8R9ZtWpVVq1ala997Wu55pprOszOmTMny5YtyxFHHJGxY8emrKwsNTU1uemmm1JTU5M77rgjpaWlHeYWLFiQuXPnprS0NKNHj07v3r1TW1ub22+/PT//+c+zaNGi9OjRo8PcT37yk1x77bVpamrKGWeckb59+2b9+vW555578vDDD6e6ujrHHHPM/v6dAAAA2A9dis1u3brlU5/6VKZOnZozzzyz3bmf/vSnmTVrVu68886cffbZGTNmTNu5VatWZdmyZamsrMzSpUszcODAJMnrr7+eqVOn5qGHHsrSpUtz+eWXt3vOurq6zJs3Lz169MjixYtz+umnJ0l27tyZq666KrW1tbntttty3XXXtZtraGjI9ddfn5aWlvzgBz/IuHHjkiR79uzJN7/5zfz0pz/NjTfemB/84Af791cCAABgv3TpO5tjx47N3/7t33YIzST59Kc/nc997nNJkgceeKDdufnz5ydJZs2a1RaaSdKnT5/Mnj07yXtXMJubm9vNLViwIC0tLfnyl7/cFppJctRRR+U73/lOSkpKsmzZsrz11lvt5hYvXpx33303f/7nf94WmklSVlaWm2++OT179szDDz+c559/vitvGwAAgANUyA2Chg0bliR59dVX2441NDTk2WefTXl5ecaPH99hZvTo0enbt29ee+21rFu3ru347t278/jjjydJLrnkkg5zAwYMyMc//vE0Njbmsccea3fu4Ycf3udcz54982d/9mft1gEAAHBoFBKbmzdvTvLe9zlbPffcc0mSk08+Od27d+90bsSIEUmS+vr6tmMvvPBC3nnnnRx99NE58cQT33eu9TWSZMeOHXnppZfane/KHAAAAMU76Nh87bXXsnz58iTJRRdd1HZ8y5YtSZLjjjtun7P9+/dvt3bvx63nOtP6nK+88kqHud69e6dnz57vO7f36wEAAFC8g4rN1hvv/Pa3v83YsWPzyU9+su3c22+/nSSd3jG21VFHHZXkvRv/7M/ckUceWdgcAAAAxTuo2Pyrv/qr1NTUpH///vmbv/mbdudaWlqSvHcn2/3x+54DAACgeAccm3/913+df/3Xf01lZWUWLVrU7vuayX9ftWy94tiZ1iuMrWu7Otd6rog5AAAAindAsfnd7343S5YsyUc+8pEsWrSo3c+atDr++OOTJFu3bt3n8zQ0NLRbu/fj3/zmN/ucaz2399wJJ5yQJHnrrbeyY8eO951rXQsAAMBakJ93AAAgAElEQVShsd+x+b3vfS8LFy7M0UcfnYULF2bIkCGdrmv9OZRNmzbl3Xff7XRNXV1dkuTUU09tOzZ48OB0794927dvb7u77P/0zDPPdJjr2bNn291rW5+3K3MAAAAUb79ic+7cubnnnnvyoQ99KAsXLswpp5yyz7X9+/fP8OHD09jYmJUrV3Y4v2bNmjQ0NKSysjIjR45sO15RUZHzzjsvSfLAAw90mHv55Zezbt26lJeX5/zzz2937oILLtjn3I4dO7J69eokyYUXXvi73ywAAAAHrMuxefvtt2fBggXp3bt3fvjDH7ZduXw/V155ZZL3IvXFF19sO75t27bMmTMnSTJjxoyUlLTfxowZM9KtW7fcfffdbVcjk/e+43ndddelubk5kyZNSu/evdvNXX755enevXt+/OMf55FHHmk7vmfPntx4443ZsWNHxo0bt8+rsQAAABSjrCuLHnnkkfz93/99kuTEE0/M0qVLO103ePDgtsBMkvHjx2fixImprq5OVVVVzjnnnJSVlaWmpqYt/C677LIOz/Oxj30sM2fOzNy5c/OlL30pY8aMSa9evVJbW5tt27bl9NNPzze+8Y0Oc/37988tt9ySa6+9Nl//+tczatSoHHvssVm/fn1eeeWVnHTSSbnpppu69IcBAADgwHUpNt988822xxs2bMiGDRs6XTd69Oh2sZkks2fPzqhRo3LvvfdmzZo1aW5uzuDBgzNhwoRMnDixw1XNVjNmzMjQoUOzcOHC1NXVZdeuXRkwYECmTJmS6dOnp6KiotO5iy++OAMGDMj8+fPz9NNPZ/369enfv3+mT5+er371q+nVq1dX3jIAAAAHoUuxeemll+bSSy894BepqqpKVVXVfs+dd955bd/f3B+nn3567rzzzv2eAwAAoBgH/DubAAAAsC9iEwAAgMKJTQAAAAonNgEAACic2AQAAKBwYhMAAIDCiU0AAAAKJzYBAAAonNgEAACgcGITAACAwolNAAAACic2AQAAKJzYBAAAoHBiEwAAgMKJTQAAAAonNgEAACic2AQAAKBwYhMAAIDCiU0AAAAKJzYBAAAonNgEAACgcGITAACAwolNAAAACic2AQAAKJzYBAAAoHBiEwAAgMKJTQAAAAonNgEAACic2AQAAKBwYhMAAIDCiU0AAAAKJzYBAAAonNgEAACgcGITAACAwolNAAAACic2AQAAKJzYBAAAoHBi8wNmd2NTu38BAAAOBbH5AVNRXpqqmfenorz0cG8FAAD4IyY2AQAAKJzYBAAAoHBiEwAAgMKJTQAAAAonNgEAACic2AQAAKBwYhMAAIDCiU0AAAAKJzYBAAAonNgEAACgcGITAACAwolNAAAACic2AQAAKJzYBAAAoHBiEwAAgMKJTQAAAApX1tWF//mf/5lf/OIXqaury4YNG7J58+a0tLTkjjvuyPjx4zud+fa3v53ly5fv8zkHDRqUlStXdnquubk51dXVue+++/LCCy+kpKQkQ4cOzaRJk3LxxRe/715XrFiR6urqbNy4Mc3NzRk0aFAmTJiQiRMnpqREXwMAABxqXY7N6urq/OM//uMBvcgZZ5yRk046qcPxysrKTtc3NTXl6quvzqOPPpqePXvm3HPPze7du1NTU5OZM2dm3bp1ueGGGzqdnTNnTpYtW5YjjjgiY8eOTVlZWWpqanLTTTelpqYmd9xxR0pLSw/ofQAAANA1XY7Nj370o5k+fXpOO+20nHbaabn++uuzZs2aLs1+4QtfyKWXXtrlTS1evDiPPvpohgwZksWLF6dPnz5Jks2bN2fy5MlZsmRJxowZk3HjxrWbW7VqVZYtW5bKysosXbo0AwcOTJK8/vrrmTp1ah566KEsXbo0l19+eZf3AgAAwP7r8mdKv/CFL+Taa6/Npz/96Zx44omHbENNTU25++67kySzZ89uC80kGThwYGbNmpUkueuuuzrMzp8/P0kya9asttBMkj59+mT27NlJkgULFqS5ufkQ7R4AAIDkD/AGQWvXrs22bdvSr1+/nHXWWR3Ojx8/PuXl5amrq8urr77adryhoSHPPvtsysvLO/0O6ejRo9O3b9+89tprWbdu3SF9DwAAAB90Xf4Y7cH41a9+lY0bN+btt9/OMccck1GjRuXcc8/t9GY99fX1SZIRI0Z0+lw9evTIkCFDUl9fn/r6+vTt2zdJ8txzzyVJTj755HTv3r3T2REjRuTVV19NfX19zjjjjCLeGgAAAJ34vcTmj3/84w7HhgwZkltvvTVDhw5td3zLli1JkuOOO26fz9e/f//U19e3rd2fub3XAgAAcGgc0o/RnnLKKbnhhhvy4IMPZu3atfnFL36R+fPn55RTTsnzzz+fK664ot1HYZPk7bffTvLeFcx9OfLII5MkO3fu3K+5o446qsMcAAAAxTukVzanTZvW7v9HHnlkjj322JxzzjmZMmVK1q1bl/nz5+fGG29sW9PS0pIk6dat23691oHOfVDtbmxKRXlph8cAAABFOCw3CKqoqMiVV16ZJHnsscfanWu9+th6pbIzreda13Z1rvWK5t5zH1QV5aWpmnl/qmbeLzQBAIDCHba70Q4ePDhJOnyM9vjjj0+SbN26dZ+zDQ0N7dYezBwAAADFO2yxuX379iQdrzIOGzYsSVJXV9fp3DvvvJNNmza1W7v3402bNuXdd9/tdLb1OU899dSD2DkAAAC/y2GLzZ/97GdJktNOO63d8ZEjR+aYY45JQ0NDamtrO8ytXLkyjY2NGTFiRNvPniTv3Wl2+PDhaWxszMqVKzvMrVmzJg0NDamsrMzIkSMLfjcAAADs7ZDFZn19fVavXp2mpqZ2x/fs2ZOFCxdmyZIlSTreRKi0tDTTp09PksyePTvbtm1rO7d58+bMmzcvSfKVr3ylw2u2fg907ty5efHFF9uOb9u2LXPmzEmSzJgxo9Pf9wQAAKA4Xb4b7bPPPtsWbEny/PPPJ0luu+22/PCHP2w7/i//8i9JkldeeSVf//rXc/TRR2fgwIHp27dvdu7cmV//+tf5r//6r5SUlGTWrFn5xCc+0eG1pk2bltra2qxevToXXXRRxo4dmz179uTJJ5/Mrl27MmXKlIwbN67D3Pjx4zNx4sRUV1enqqoq55xzTsrKylJTU5MdO3Zk3Lhxueyyy7r+1wEAAOCAdDk2d+zYkfXr13c4vnnz5k7XDx06NFOnTk1dXV1eeeWVPPfcc+nWrVv69euXSy+9NJMnT+7wEdpWpaWlufPOO7Ns2bL86Ec/yhNPPJGSkpIMHz48kyZNSlVV1T73OXv27IwaNSr33ntv1qxZk+bm5gwePDgTJkzIxIkTXdUEAAD4PehybJ599tnZuHFjl594wIABuf766w9oU0lSUlKSyy677ICuRFZVVb1vkAIAAHBoucwHAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTm2R3Y1O7fwEAAA6W2CQV5aWpmnl/KspLD/dWAACAPxJiEwAAgMKJTQAAAAonNgEAACic2AQAAKBwYhMAAIDCiU0AAAAKJzYBAAAonNgEAACgcGITAACAwolNAAAACic2AQAAKJzYBAAAoHBiEwAAgMKJTQAAAAonNgEAACic2AQAAKBwYhMAAIDCiU0AAAAKJzYBAAAonNgEAACgcGITAACAwolNAAAACic2AQAAKJzYBAAAoHBiEwAAgMKJTQAAAAonNmmzu7Gp08cAAAD7q+xwb4A/HBXlpamaeX+SZMW8zx7m3QAAAP+bubIJAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJp1q/ekTP4ECAAAcCLFJp1p/BqWivPRwbwUAAPhfSGwCAABQOLEJAABA4cQmAAAAhRObAAAAFE5sAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sQkAAEDhxCYAAACFK+vqwv/8z//ML37xi9TV1WXDhg3ZvHlzWlpacscdd2T8+PHvO7tixYpUV1dn48aNaW5uzqBBgzJhwoRMnDgxJSX77t3HH388ixYtyoYNG7Jr164MGDAgn/nMZzJ9+vRUVFTsc279+vX5h3/4hzz99NPZsWNH+vfvn3HjxuWrX/1qevXq1dW3DAAAwAHqcmxWV1fnH//xH/f7BebMmZNly5bliCOOyNixY1NWVpaamprcdNNNqampyR133JHS0tIOcwsWLMjcuXNTWlqa0aNHp3fv3qmtrc3tt9+en//851m0aFF69OjRYe4nP/lJrr322jQ1NeWMM85I3759s379+txzzz15+OGHU11dnWOOOWa/3wcAAABd1+XY/OhHP5rp06fntNNOy2mnnZbrr78+a9ased+ZVatWZdmyZamsrMzSpUszcODAJMnrr7+eqVOn5qGHHsrSpUtz+eWXt5urq6vLvHnz0qNHjyxevDinn356kmTnzp256qqrUltbm9tuuy3XXXddu7mGhoZcf/31aWlpyQ9+8IOMGzcuSbJnz55885vfzE9/+tPceOON+cEPftDVtw0AAMAB6PJ3Nr/whS/k2muvzac//emceOKJXZqZP39+kmTWrFltoZkkffr0yezZs5O8dwWzubm53dyCBQvS0tKSL3/5y22hmSRHHXVUvvOd76SkpCTLli3LW2+91W5u8eLFeffdd/Pnf/7nbaGZJGVlZbn55pvTs2fPPPzww3n++ee7+rYBAAA4AIfsBkENDQ159tlnU15e3ul3OkePHp2+ffvmtddey7p169qO7969O48//niS5JJLLukwN2DAgHz84x9PY2NjHnvssXbnHn744X3O9ezZM3/2Z3/Wbh0AAACHxiGLzeeeey5JcvLJJ6d79+6drhkxYkSSpL6+vu3YCy+8kHfeeSdHH330Pq+gts61vkaS7NixIy+99FK7812ZAwAAoHiHLDa3bNmSJDnuuOP2uaZ///7t1u79uPVcZ1qf85VXXukw17t37/Ts2fN95/Z+PQAAAIp3yGLz7bffTpJO7xjb6qijjkry3o1/9mfuyCOPLGwOAACA4h2y2GxpaUmSdOvW7Q96DgAAgOIdsthsvWrZesWxM61XGFvXdnWu9VwRcwAAABTvkMXm8ccfnyTZunXrPtc0NDS0W7v349/85jf7nGs9t/fcCSeckCR56623smPHjveda10LAADAoXHIYnPYsGFJkk2bNuXdd9/tdE1dXV2S5NRTT207Nnjw4HTv3j3bt29vu7vs//TMM890mOvZs2fb3Wtbn7crcwAAABTvkMVm//79M3z48DQ2NmblypUdzq9ZsyYNDQ2prKzMyJEj245XVFTkvPPOS5I88MADHeZefvnlrFu3LuXl5Tn//PPbnbvgggv2Obdjx46sXr06SXLhhRce8PsCAADgdztksZkkV155ZZJk7ty5efHFF9uOb9u2LXPmzEmSzJgxIyUl7bcxY8aMdOvWLXfffXfb1cjkve94XnfddWlubs6kSZPSu3fvdnOXX355unfvnh//+Md55JFH2o7v2bMnN954Y3bs2JFx48ZlyJAhhb9XAAAA/ltZVxc+++yzbYGYJM8//3yS5LbbbssPf/jDtuP/8i//0vZ4/PjxmThxYqqrq1NVVZVzzjknZWVlqampaQu/yy67rMNrfexjH8vMmTMzd+7cfOlLX8qYMWPSq1ev1NbWZtu2bTn99NPzjW98o8Nc//79c8stt+Taa6/N17/+9YwaNSrHHnts1q9fn1deeSUnnXRSbrrppq6+ZQAAAA5Ql2Nzx44dWb9+fYfjmzdvft+52bNnZ9SoUbn33nuzZs2aNDc3Z/DgwZkwYUImTpzY4apmqxkzZmTo0KFZuHBh6urqsmvXrgwYMCBTpkzJ9OnTU1FR0encxRdfnAEDBmT+/Pl5+umns379+vTv3z/Tp0/PV7/61fTq1aurb5kkuxubUlFe2uExAADA++lybJ599tnZuHHjAb1IVVVVqqqq9nvuvPPOa/v+5v44/fTTc+edd+73HB1VlJemaub9SZIV8z57mHcDAAD8b3FIv7MJAADAB5PYBAAAoHBiEwAAgMKJTQAAAAonNumy3Y1NnT4GAAD4n7p8N1pwZ1oAAKCrXNkEAACgcGITAACAwolNAAAACic2AQAAKJzYBAAAoHBiEwAAgMKJTQAAAAonNgEAACic2AQAAKBwYpMDsruxqd2/AAAAexObHJCK8tJUzbw/FeWlh3srAADAHyCxCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4sUlh3KEWAABoJTYpjDvUAgAArcQmAAAAhRObAAAAFE5sAgAAUDixyUFxMyAAAKAzYpOD0npToKqZ9x/urQAAAH9AxCaF2/tqpyufAADwwVR2uDfAH5/Wq51JsmLeZw/zbgAAgMPBlU0AAAAKJzYBAAAonNgEAACgcGITAACAwolNAAAACic2AQAAKJzYBAAAoHBiEwAAgMKJTQ6p3Y1N7f4FAAA+GMQmh1RFeWmqZt6fivLSw70VAADg90hsAgAAUDixCQAAQOHEJgAAAIUTmwAAABRObAIAAFA4scnvxd4/feJnUAAA4I9f2eHeAB8MrT+BkiQr5n32MO8GAAA41FzZBAAAoHBiEwAAgMKJTQAAAAonNgEAACic2AQAAKBwYhMAAIDCiU0AAAAKd8h/Z/Pb3/52li9fvs/zgwYNysqVKzscb25uTnV1de6777688MILKSkpydChQzNp0qRcfPHF7/uaK1asSHV1dTZu3Jjm5uYMGjQoEyZMyMSJE1NSoq8BAAAOtUMem63OOOOMnHTSSR2OV1ZWdjjW1NSUq6++Oo8++uj/b+/+o6Ku8j+OvwBBxJ+lCGgWkgIGZGiR4rHdLfKQRT+wtvyZRXr6sdu2R9b2bB4Dt7a2NNOtlDVTU+Qcj5qbmriappYcoVUUlVVJMX+EsaQZ/hxgvn94Zr4gMzAMn4GZ4fn4R/x87r1zP3cud3jPvZ/7UadOnTRs2DBdvXpVeXl5mjJligoLCzVt2jSbr5OZmanly5erffv2Gjp0qNq1a6e8vDzNmDFDeXl5mjNnjvz8/Ay/PgAAAADA/2uxYPOJJ55QamqqQ2mXLFmiLVu2qF+/flqyZIl69OghSSotLdXYsWO1dOlSDRkyRElJSXXybdy4UcuXL1dwcLCWLVum8PBwSdL//vc/TZgwQZs2bdKyZcv09NNPG3ptaJqrpmoF+PtZ/wUAAADgfdxuTWl1dbU+/vhjSVJGRoY10JSk8PBwpaenS5Lmz59fL29WVpYkKT093RpoSlKPHj2UkZEhSVqwYIFqampcVHs4IsDfTylT/kWgCQAAAHgxtws29+zZo4qKCoWGhuquu+6qdz45OVn+/v4qKirSmTNnrMfLysp04MAB+fv7Kzk5uV6+hIQEhYSEqLy8XIWFhS69BgAAAABo61psGe2uXbt06NAhXbx4Ud27d9fgwYM1bNiwehv2FBcXS5Li4uJsltOhQwf169dPxcXFKi4uVkhIiCTp4MGDkqT+/fsrMDDQZt64uDidOXNGxcXFGjRokFGXBgAAAAC4TosFm2vWrKl3rF+/fnrvvfcUFRVlPXby5ElJUq9eveyWFRYWpuLiYmvapuSrnRYAAAAA4BouX0YbHR2tadOmaf369dqzZ4927NihrKwsRUdHq6SkRM8880yd5bAXL16UdG0G056goCBJ0oULF5qUr2PHjvXyofVcNVXb/BkAAACA53P5zObEiRPr/D8oKEg9e/ZUYmKixo8fr8LCQmVlZWn69OmSJLPZLEny8fFp0us4mw+tx7JRkCStnfVIK9cGAAAAgJFabYOggIAATZ48WZK0bds263HL7KNlptIWyzlLWkfzWWY0a+cDAAAAABivVXejjYiIkKQ6y2h79+4tSTp9+rTdfGVlZXXSNicfAAAAAMB4rRpsnjt3TlLdmcbbbrtNklRUVGQzz6VLl3TkyJE6aWv/fOTIEV2+fNlmXkuZAwYMaGbNYTTLPZvcuwkAAAB4h1YNNjds2CBJio2NtR6Lj49X9+7dVVZWpoKCgnp5cnNzZTKZFBcXZ33siXRtp9mYmBiZTCbl5ubWy5efn6+ysjIFBwcrPj7eBVeD5rDcvxng79faVQEAAABgAJcGm8XFxdq6dauqq+vOVlVVVWnRokVaunSppLqbCPn5+SktLU2SlJGRoYqKCuu50tJSzZo1S5L0/PPP13s9yz2gM2fO1PHjx63HKyoqlJmZKUmaNGlSvWd7AgAAAACM5dLdaE+dOqWXXnpJ3bp1U3h4uEJCQnThwgUdPnxYP/74o3x9fZWenq7hw4fXyTdx4kQVFBRo69atGjFihIYOHaqqqirt3LlTV65c0fjx45WUlFTv9ZKTkzV69Gjl5OQoJSVFiYmJateunfLy8lRZWamkpCSNGzfOlZcMAAAAAJCLg82oqChNmDBBRUVFOnXqlA4ePCgfHx+FhoYqNTVVY8eOrbOE1sLPz08fffSRli9frtWrV+vrr7+Wr6+vYmJiNGbMGKWkpNh9zYyMDA0ePFjZ2dnKz89XTU2NIiIiNGrUKI0ePZpZTQAAAABoAS4NNvv06aPXXnvNqby+vr4aN26cUzORKSkpDQakcF9XTdXW+zZr/wwAAADAs7g02ASayrJRkCStnfVIK9cGAAAAgLNYUwoAAAAAMBzBJgAAAADAcASbAAAAAADDEWwCAAAAAAxHsAm3ddVUXedfAAAAAJ6DYBNuy7IzLY8/AQAAADwPwSYAAAAAwHAEmwAAAAAAwxFsAgAAAAAMR7AJt1d7gyA2CwIAAAA8Q7vWrgDQGMtGQZK0dtYjrVwbAAAAAI5gZhMehVlOAAAAwDMwswmPwiwnAAAA4BmY2YTHssxsMsMJAAAAuB+CTXgsyyxngL9fa1cFAAAAwHUINgEAAAAAhiPYhFdhaS0AAADgHgg24VVYWgsAAAC4B4JNeDxmMQEAAAD3Q7AJj2eZzbQ8EkXieZwAAABAa+M5m/BKPI8TAAAAaF3MbAIAAAAADEewCQAAAAAwHMEmvB6PQwEAAABaHsEmvB6PQwEAAABaHsEm2iRmOwEAAADXIthEm8RsJwAAAOBaBJto03geJwAAAOAaBJtoM2wFk5YZTsszOe2lAwAAANA0BJtoM2wFlrbOs7QWAAAAaD6CTeA6LK0FAAAAmq9da1cAcDeWGU5JWjvrkVauDQAAAOCZmNkEmoiZTwAAAKBxBJuAA2oHlbXv/eT+TgAAAMA2gk3AAY1tLgQAAACgLoJNoAEskwUAAACcQ7AJNKD2ZkG2WILRK9zHCQAAANRBsAk0gyUYbc99nAAAAEAdBJsAAAAAAMMRbAIGsyyjZTktAAAA2jKCTcBglqW1tZfTEoACAACgrSHYBFqArY2GrhCAAgAAwIsRbAIuYiuIrP28zvY2ZkABAAAAb0GwCbhI7cDSUbaW2/J4FQAAAHgigk2gldUOHG3d78njVQAAAOCJCDaBVmZrBrSxmUtHZzttzZDaOgYAAAAYjWATcEONLcG1NdtZmyWIDLAxG2rrGAAAAGA0gk3ASzQWoDKLCQAAgJZEsAl4IVuPWrF1zBaeCQoAAAAjEGwCqMPWJkUAAABAUxFsAm2YrY2GHEl//c8AAADA9Qg2gTasoY2GageTVxrYcMgdAlB3qAMAAADqItgEYFPtwLJ9A/d72tvd1tHHs9g619SAkR12AQAA3A/BJgCnNPZcz4ZmTa80Mmva0GsAAADAM3htsLl27VqNGTNGgwcPVnx8vFJTU5Wdna2amprWrhrgFRra8bYpzwe1NWtqq+zGgtvrjxGoAgAAtC6vDDYzMzOVnp6u/fv3684771RiYqJKS0s1Y8YMvfzyy6qu5o9QwBM0do+oo494sbVE19YxR5f8AgAAoHFeF2xu3LhRy5cvV3BwsD7//HNlZWXpww8/1L///W/deuut2rRpk5YtW9ba1QTQRFcbWG5ri710DQWo7RsJbq/YCEptHSNQBQAA8MJgMysrS5KUnp6u8PBw6/EePXooIyNDkrRgwQKW0wIexlaQ2NS8tfM3FrTaCm7bN7AM2NF7U2uX7Wgg60zw2thy4qYuN3a3HX9ZLg0AgPvzqmCzrKxMBw4ckL+/v5KTk+udT0hIUEhIiMrLy1VYWNgKNQTgKYwIbm0FpfYC3oYC2docnV21lG1rh+DGzje2gZOjs74N3VPb3EDc0fo7swMygSwAAMbwqmDz4MGDkqT+/fsrMDDQZpq4uDhJUnFxcYvVCwBqa2og25zZVVsBZmPnGwt4HZ31tbAVJDY3EHe0/rbq0NAOyLWP12bEcunGZocdDd4bC4ybMyPeElxRV3f9gqA594M7+n57OkfvqbeVx5vaAfBW7Vq7AkY6efKkJKlXr15204SFhdVJ2xjLZkJlZWXNrJ1rmC7+JOna9Zgu/mT915XHWvr1qIPn1csd6uCu9Wrp1/vxzA967s1NkqSPX7vfofONlW3JY6s8e+kceY2mlO1o/e3V4frXsFXO9ceuL2feq/fJv52vrlbVKKDdte9uLT/bOiapwby1zzf0erbKsZW39rmG6tWU+htVTkN1NVXVyL+dr/VfR+pgKdPoa659zNF61T5vqZet97F2OsvPjV1T7fIsbNXHldfszPvTlP5Q+zprt5ej7/f17X99O9k672gbNtYODR1rrI728ti7DmfqZa9sW/2vqX3EUU39/bGXv7FrtvV6Ta1rY3Uwqh2MqperWWKhpm606mM2m82uqFBrmD9/vmbPnq2UlBTNnDnTZprZs2dr/vz5evLJJzVjxoxGy/z22281duxYo6sKAAAAAB4lOztbd955p8PpvWpm0xI3+/j4GFZmbGyssrOzFRwcLD8/v8YzAAAAAIAXqa6uVnl5uWJjY5uUz6uCzY4dO0qSLl68aDfNhQsX6qRtTGBgYJOidwAAAADwNrfcckuT85fFuQcAABXOSURBVLj/AuEm6N27tyTp9OnTdtNY1htb0gIAAAAAjOdVweZtt90mSTpy5IguX75sM01RUZEkacCAAS1WLwAAAABoa7wq2AwLC1NMTIxMJpNyc3Prnc/Pz1dZWZmCg4MVHx/fCjUEAAAAgLbBq4JNSZo8ebIkaebMmTp+/Lj1eEVFhTIzMyVJkyZNkq+v1106AAAAALgNr3r0iUVGRoZycnLUvn17JSYmql27dsrLy1NlZaWSkpI0d+5cdpYFAAAAABfyymBTktauXavs7GwdPnxYNTU1ioiI0KhRozR69GhmNQEAAADAxbw22AQAAAAAtB6ves6mu1u7dq1ycnJ06NAh1dTUqG/fvs2abd2+fbsWL16s/fv368qVK+rTp48efPBBpaWlKSAgwG6+vXv36p///Kd2796tyspKhYWFKSkpSS+88II6d+7cnEt0W0a0fU1NjQoLC7Vt2zYVFBTo5MmT+umnn9SlSxfFxMToySefVFJSks28//jHP/TBBx/YLTsgIMC6U7K3MarfN7cNjf798wRGXPPJkyd13333OZR22bJluuuuu6z/b4v9/ujRo9qxY4eKioq0f/9+lZaWymw2a86cOUpOTna6XGffS2c/JzyVke1vMpn07bffatu2bdq9e7dOnz6tc+fO6YYbblB8fLzGjh2ru+++22beP//5z/rss8/slt23b1+bGxl6MqP7fnPasKamRjk5OVq1apWOHTsmX19fRUVFacyYMXrooYeaXBd3Z2Tb79q1SxMmTHAo7datW9WrVy/r/9tav2/OGNEYbxrzCTZbSGZmppYvX6727dtr6NCh1vtIZ8yYoby8PM2ZM6dJ95EuWLBAM2fOlJ+fnxISEtSlSxcVFBTo/fff11dffaXFixerQ4cO9fKtW7dOU6dOVXV1tQYNGqSQkBDt3btXCxcu1ObNm5WTk6Pu3bsbeemtzqi2P3HihEaPHi1J6tatm2JjY9WlSxedOHFC27dv1/bt25Wamqq//e1v8vHxsVlGdHS0zcfutGvnnb+KRvd7ybk2dEU93J1R1xwUFKTHHnvM7vmSkhIVFRWpY8eOiomJsZmmLfX7nJwcffrpp4aW6ex76eznhCczsv0LCgr0zDPPSJKCg4MVExOjDh066LvvvtPGjRu1ceNGvfjii/rDH/5gt4xBgwbZfAh6cHCwIXV0J67o+1LT27C6ulq/+93vtGXLFnXq1EnDhg3T1atXlZeXpylTpqiwsFDTpk0zvJ6tyci279GjR4Nj/r59+/Tdd9/p5ptvVlhYmM00baXfGzFG2OJ1Y74ZLpebm2uOjIw0Dxs2zHzs2DHr8fLycvMDDzxgjoyMNC9evNjh8vbt22eOiooyDxw40FxYWGg9XllZaR47dqw5MjLS/Oabb9bL98MPP5hvv/12c3R0tHnTpk3W4yaTyfzKK6+YIyMjzS+++KJzF+mmjGz748ePmydMmGDetm2buaqqqs65Xbt2me+44w5zZGSkeeXKlfXyzp071xwZGWmeO3dus67Hkxjd751tQ6Pr4Qla8pqfe+45c2RkpHnatGn1zrXFfr9ixQrz3//+d/P69evNx48fN48bN84cGRlp3rBhg1PlOfteOvs54emMbP+dO3eaf//735sLCgrqnVu/fr15wIAB5sjISHNeXl6986+++qo5MjLSvGrVKqeuwxMZ3fedbcOFCxeaIyMjzSNHjjSXl5dbjx87dsycmJhojoyMrPM3kDcwuu0bMnLkSHNkZKR53rx59c61tX7fnDHCHm8c871z7ZibycrKkiSlp6crPDzcerxHjx7KyMiQdO3biJqaGofKW7Bggcxms5577jkNHDjQerxjx45666235Ovrq+XLl+v8+fN18i1ZskSXL1/Wo48+Wme5Z7t27fTXv/5VnTp10ubNm1VSUuLklbofI9v+5ptv1pIlS3TPPffU+0YpISFBkyZNkiR9/vnnxlTewxnd7z29Hi2ppa75zJkz+uabbyRJjz/+eLPK8hZPPPGEpk6dqpEjR+rmm29udnnOvpfOfk54OiPbf+jQoZo7d67uvPPOeudGjhxpnf1hzL/G6L7vjOrqan388ceSrj2ZoEePHtZz4eHhSk9PlyTNnz+/VernKi3V9nv27FFJSYn8/PwanP1sK1wxRnjjmE+w6WJlZWU6cOCA/P39ba6bT0hIUEhIiMrLy1VYWNhoeVevXtX27dslSQ8//HC983369NEdd9whk8mkbdu21Tm3efNmu/k6deqk3/zmN3XSeTqj274xt912m/V127qWbnt3r0dLaslrXr16taqrq9W/f/86H24whrPvZXM+J+A4y5h/5syZVq4JLPbs2aOKigqFhobWuX/cIjk5Wf7+/ioqKuJ9c8KqVaskScOHD1dISEgr18b9NXWM8NYx3/tumHEzBw8elCT1799fgYGBNtPExcXpzJkzKi4u1qBBgxos79ixY7p06ZK6detm99uruLg47d69WwcPHlRKSookqbKyUt9//731vL18a9eutdbZ0xnd9o0pLS2VJPXs2dNumgMHDujdd9/V+fPn1bVrVw0cOFC/+tWvvG6jDle2fVPasKX7gDtoyWu2bATR2KxmW+n3RnP2vXT2cwJNYxnzG7oPbdeuXTp06JAuXryo7t27a/DgwRo2bJjXbkrmCk1pw+LiYkn2/87p0KGD+vXrp+LiYhUXFxMwNcGlS5f0xRdfSGp8zKffX+PIGFGbt475BJsudvLkSUmqs1vX9Sw3WFvSOlKevZuya7/WqVOn6uXr0qWLOnXq1GA+R+rhCYxu+4ZcunRJS5culSSNGDHCbrqtW7dq69atdY6Fhobq3XffVUJCQrPq4E5c2fZNacOW7APuoqWuOT8/X8ePH5e/v7/Nb1Jrayv93mjOvpfOfk7AceXl5dYvWxoa89esWVPvWL9+/fTee+8pKirKZfXzJk1pQ0d/Z4qLi71mzG8pubm5unDhgrp3765f//rXDaal3zs+RtTmrWN+2/qKoRVcvHhRkhrc/aljx46SpAsXLhhSXlBQUL3ynM3nyYxu+4ZkZmbq5MmT6tevn5588sl65/v06aMpU6boX//6l/7zn/8oLy9PS5YsUUJCgsrKyjR58mT997//bVYd3Ikr2t6ZNmzJPuAuWuqaLcup7r33Xt14440207S1fm80Z9/Ltjjet6Sqqir96U9/0i+//KKhQ4fq3nvvrZcmOjpa06ZN0/r167Vnzx7t2LFDWVlZio6OVklJiZ555hmWcTbCmTak77uOZcx/5JFH5O/vbzMN/f4aR8YIW7x1zGdm08XMZrMk2X0URkuVZ3Q9PEFLXfOHH36ozz77TJ07d9b7779vc2ngo48+Wu/YkCFDNGTIEL388svauHGjZs+ebb0x3NO5ou2daUP6vWtUVlZq48aNkqRRo0bZTdfW+r3RGO/d0+uvv668vDyFhYXp3XfftZlm4sSJdf4fFBSknj17KjExUePHj1dhYaGysrI0ffr0FqixZ3KmDen7rnH8+HEVFBRIangJLf3+GkfGCFu8dcxnZtPFLN9AWL51sMXyLYMlbXPLs5yrXZ6z+TyZ0W1vy6JFizR37lwFBQVpwYIF6t+/f5PLePHFFyVJ33zzjUwmk1P1cDct0fa12WvDlq6HO2iJa16/fr0uXbqk0NBQDR8+3KkyvLHfG83Z97Itjvct5Y033tDKlSsVHBysxYsXN/m5gQEBAZo8ebIksTmTkxpqQ/q+a1hmNePj43Xrrbc2OX9b6vfNGSO8dcwn2HSx3r17S5JOnz5tN41l91JLWkfK++GHH+ymsZyrXd5NN90kSTp//rwqKysbzGdJ6+mMbvvrLV26VG+//bYCAwOVlZWl+Ph4p+oZEREhSTKZTDp79qxTZbgbV7f99ey1YUvXwx20xDVb/vB47LHHnN7wwRv7vdGcfS+d/ZxAw95++20tXbpUN954oxYvXlznsQRNYen7bWE5oavYa8O2OOa7WnV1tfUezIZWsjSmLfT75o4R3jrmE2y6mGXb4yNHjujy5cs20xQVFUmSBgwY0Gh5ERERCgwM1Llz56y7y15v37599crr1KmTdYcqy+s5ks+TGd32tWVnZ+uNN95Q+/btNW/evGZtcnLu3Dnrz5Y19Z7OlW1vi702bOl6uANXX3NJSYn27t0rHx+fZv3h4Y393mjOvpfOfk7AvnfeeUeLFi1St27dtGjRIvXr18/psix9n5k159lrQ8vvjL2/cy5duqQjR47USYuGff311zpz5oyCgoI0cuRIp8vx9n5vxBjhrWM+waaLhYWFKSYmRiaTSbm5ufXO5+fnq6ysTMHBwQ7NjAUEBOiee+6RZPshsSdOnFBhYaH8/f3r7RZ233332c1XWVlp3S3y/vvvb7QensDotrfIycnRjBkzFBAQoA8//FCJiYnNqueGDRskSX379rW7U7CncVXb22OvDVu6Hu7A1de8cuVKSdLdd9+tPn36OF1Pb+z3RnP2vWzO5wTqmzlzphYuXKiuXbtq0aJFio6OblZ5lr4fGxtrRPXaJHttGB8fr+7du6usrMx6j2Ftubm5MplMiouL47EnDrKM+Q888ECzAkVv7vdGjRHeOuYTbLYAyzr1mTNn6vjx49bjFRUVyszMlCRNmjSpznK0ZcuWKTk5WVOnTq1X3qRJk+Tj46OPP/7Y+k2FdG0d91/+8hfV1NRozJgx6tKlS518Tz/9tAIDA7VmzRp9+eWX1uNVVVWaPn26KisrlZSU1KxvbN2N0W2/YsUKZWZmKiAgQB988IFD96udPn1aa9eu1dWrV+scN5vNWrNmjd577z1J9W+s93RGtn1z2tCZeng6o/u9hclksn6QNfactbba750xa9YsJScna9asWfXOOdt/nf2caIsaav/3339fCxYsUJcuXfTJJ584NBtWXFysrVu3qrq6us7xqqoqLVq0yPqYLPq+/bZ3tg39/PyUlpYmScrIyFBFRYX1XGlpqfV1nn/+eaMvxeM01O8tfvrpJ+tERGNjflvt986MEW1tzGc32haQnJys0aNHKycnRykpKUpMTFS7du2Ul5dnDfDGjRtXJ8/Zs2d17NgxmzcW33777ZoyZYpmzpypp556SkOGDFHnzp1VUFCgiooKDRw4UH/84x/r5QsLC9Obb76pqVOn6qWXXtLgwYPVs2dP7d27V6dOndItt9yiGTNmuKwdWoORbV9cXKzp06fLbDbrpptu0oYNG6zf1NV2ww036NVXX7X+/+eff1Z6erpef/119e3bV7169ZLJZNKRI0esz0YaN26cnnrqKRe0QOsxsu2b04bO1MPTGT3mWHz11VeqqKhQly5dGn1uWFvt9wcOHLD+QSBdW3YsSbNnz9Ynn3xiPb5ixQrrz+Xl5Tp27JjKy8vrleds/3X2c8LTGdn+X375pebNmydJuvnmm7Vs2TKbrxkREWH9A1G69hy7l156Sd26dVN4eLhCQkJ04cIFHT58WD/++KN8fX2Vnp7u9OZa7srItm9OG06cOFEFBQXaunWrRowYoaFDh6qqqko7d+7UlStXNH78eCUlJRl9+a3K6HHH4vPPP5fJZFJERIQGDRrUYB3aYr93doxoa2M+wWYLycjI0ODBg5Wdna38/HzV1NQoIiJCo0aN0ujRo5s8qzJp0iRFRUVp0aJFKioq0pUrV9SnTx+NHz9eaWlpNh+/IUkPPfSQ+vTpo6ysLO3evVt79+5VWFiY0tLS9MILL6hz585GXK5bMartz58/b91e+ujRozp69KjNdL17964TbIaGhiotLU1FRUX6/vvvVVJSopqaGgUHB2vkyJH67W9/q6FDhzb/Qt2QUW3f3DY0+vfPE7jimi0bAz300ENq3759g2nbar+vrKzU3r176x0vLS11ukxn30tnPyc8mZHt//PPP1t/3r9/v/bv328zXUJCQp0/JKOiojRhwgQVFRXp1KlTOnjwoHx8fBQaGqrU1FSNHTvWK5cSGtn2zWlDPz8/ffTRR1q+fLlWr16tr7/+Wr6+voqJidGYMWOUkpLS5Pq4O1eMO9L/j/mO3J/fFvu9s2NEY7xtzPcxW/56BgAAAADAIN73dT4AAAAAoNURbAIAAAAADEewCQAAAAAwHMEmAAAAAMBwBJsAAAAAAMMRbAIAAAAADMdzNgEAAADADRw9elQ7duxQUVGR9u/fr9LSUpnNZs2ZM0fJycmGvc748eOVn5/faLrU1FS99dZbTr8OwSYAAAAAuIGcnBx9+umnLn+d4cOHq3fv3jbPmUwmrVu3TpJ09913N+t1CDYBAAAAwA1ERkYqLS1NsbGxio2N1WuvvebQDGRTTZ482e65L774QuvWrVPnzp2bPZtKsAkAAAAAbuCJJ55o7Spo1apVkqQHH3xQgYGBzSqLYBMAAAAAPNjFixeVnZ2t3NxcHTt2TFVVVerTp4+Sk5P17LPPqmPHjg6V88MPP2jnzp2SpMcff7zZ9SLYBAAAAAAPVVZWprS0NJWUlOjGG29UfHy8AgICVFRUpA8++ECbNm3S0qVL1bVr10bLWr16tWpqahQVFaW4uLhm141HnwAAAACABzKbzXrllVdUUlKicePGacuWLVq4cKHmzZunzZs36+GHH9ahQ4cc2lHWbDbrs88+k2TMrKZEsAkAAAAAHmn79u3as2eP7rjjDr322mvq0KGD9VxgYKAyMzPVvXt3rV27Vj///HODZe3atUsnTpxQQECAUlJSDKkfwSYAAAAAeKDt27dLkkaMGCFf3/qhXVBQkGJjY1VVVaWioqIGy1q5cqUk6b777tMNN9xgSP24ZxMAAAAAPNCJEyckSe+8847eeeedBtP+9NNPds/98ssv2rRpkyTjltBKBJsAAAAA4JGqq6slSQkJCerdu3eDaXv16mX33Lp163T58mX16tVLiYmJhtWPYBMAAAAAPFBYWJgkKTk5WWPHjnW6HMuzNVNTU20ux3UW92wCAAAAgAe65557JEm5ublOl3H48GEVFRXJx8dHqampRlVNEsEmAAAAAHikpKQkxcTEKD8/X9OnT9e5c+fqpTlx4oSys7PtlmHZGCgxMbHRpbhN5WM2m82GlggAAAAAaLIDBw4oMzPT+v+SkhJduHBB4eHh6tq1q/X4ihUrrD+XlZVp0qRJOnz4sDp27Kjo6GiFhobq7NmzOn36tEpLS9WjRw9988039V7PZDJp+PDhOnv2rGbPnq2RI0caej3cswkAAAAAbqCyslJ79+6td7y0tNRuntDQUK1cuVIrV67Uhg0bdPjwYe3bt0/dunVTz5499eyzz+r++++3mXfLli06e/asunXrpqSkJKMuw4qZTQAAAACA4bhnEwAAAABgOIJNAAAAAIDhCDYBAAAAAIYj2AQAAAAAGI5gEwAAAABgOIJNAAAAAIDhCDYBAAAAAIYj2AQAAAAAGI5gEwAAAABgOIJNAAAAAIDh/g9NbrCvyluN1AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,10))\n",
    "hist_views = plt.hist(df['views'],300)\n",
    "fig_views = plt.gcf()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see, the distribution is definitely skewed right, with a smooth (but not linear) decreasing trend. Most videos has a small number of views and very few videos have a view over 10,000,000 times."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Shape & Central Tendency & Variability"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To get a more precise (numerical) idea of the distribution of the view data, we can look at some statistics of the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    3.871500e+04\n",
       "mean     1.647253e+06\n",
       "std      2.763398e+06\n",
       "min      5.490000e+02\n",
       "25%      2.417205e+05\n",
       "50%      6.626880e+05\n",
       "75%      1.718966e+06\n",
       "max      1.999988e+07\n",
       "Name: views, dtype: float64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['views'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7.63636671232e+12\n"
     ]
    }
   ],
   "source": [
    "# Variance for views\n",
    "var_views = df['views'].var()\n",
    "print(var_views)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As the statictics shows, the video with the least view has 549 views, and the video with the most view has about 200,000,000 views. The range of the data is huge, and the standard error is even larger than the mean. This confirms that the distribution of the data is skewed, and shows that the data are sparsely distributed."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Correlation Heatmap"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we can use a heatmap to show the correlation between different columns in the dataframe (but only numerical data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1d6e7752eb8>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "corr = df.corr()\n",
    "sns.heatmap(corr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the heatmap we can see that comment_count and likes are strongly correlated, as well as views and likes. This could be explained by the fact that when people like a video, they are more likely to leave a comment. Also, due to the nature of the recommendation algorithm of Youtube, usually people only click on and watch the videos that they are more likely to enjoy. This is further supported by the fact that the relation between views and dislikes are not as strong, because Youtube wouldn't want to recommend videos that people don't like"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analysis I: Linear Regression on Number of Tags"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Does more tags means more views? To prove that there is a positive correlation between number of tags and number of views, we chose to do a linear regression. But first we need to retreive the number of tags for each video.\n",
    "\n",
    "## Counting the number of tags per Video\n",
    "Since tags in the dataframe are 'string' data types, we first thought about comparing the lengths of the strings and use the idea that longer string means a higher number of tags. However, we then discovered that there are separations signs ('|') between the tags. Therefore, we can actually just find out how many of these separations there are and add two to it, to determine the amount of tags of each video. Videos without tags are already removed so we don't need to worry about them"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "number_of_tags = df['tags'].str.find('|') + 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Add this information to our main dataframe and draw a histogram to look at the distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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>trending_date</th>\n",
       "      <th>title</th>\n",
       "      <th>publish_time</th>\n",
       "      <th>tags</th>\n",
       "      <th>views</th>\n",
       "      <th>likes</th>\n",
       "      <th>dislikes</th>\n",
       "      <th>comment_count</th>\n",
       "      <th>number_of_tags</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>we want to talk about our marriage</td>\n",
       "      <td>2017-11-13</td>\n",
       "      <td>shantell martin</td>\n",
       "      <td>748374</td>\n",
       "      <td>57527</td>\n",
       "      <td>2966</td>\n",
       "      <td>15954</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>the trump presidency: last week tonight with j...</td>\n",
       "      <td>2017-11-13</td>\n",
       "      <td>last week tonight trump presidency|\"last week ...</td>\n",
       "      <td>2418783</td>\n",
       "      <td>97185</td>\n",
       "      <td>6146</td>\n",
       "      <td>12703</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>racist superman | rudy mancuso, king bach &amp; le...</td>\n",
       "      <td>2017-11-12</td>\n",
       "      <td>racist superman|\"rudy\"|\"mancuso\"|\"king\"|\"bach\"...</td>\n",
       "      <td>3191434</td>\n",
       "      <td>146033</td>\n",
       "      <td>5339</td>\n",
       "      <td>8181</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>nickelback lyrics: real or fake?</td>\n",
       "      <td>2017-11-13</td>\n",
       "      <td>rhett and link|\"gmm\"|\"good mythical morning\"|\"...</td>\n",
       "      <td>343168</td>\n",
       "      <td>10172</td>\n",
       "      <td>666</td>\n",
       "      <td>2146</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-11-14</td>\n",
       "      <td>i dare you: going bald!?</td>\n",
       "      <td>2017-11-12</td>\n",
       "      <td>ryan|\"higa\"|\"higatv\"|\"nigahiga\"|\"i dare you\"|\"...</td>\n",
       "      <td>2095731</td>\n",
       "      <td>132235</td>\n",
       "      <td>1989</td>\n",
       "      <td>17518</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  trending_date                                              title  \\\n",
       "0    2017-11-14                 we want to talk about our marriage   \n",
       "1    2017-11-14  the trump presidency: last week tonight with j...   \n",
       "2    2017-11-14  racist superman | rudy mancuso, king bach & le...   \n",
       "3    2017-11-14                   nickelback lyrics: real or fake?   \n",
       "4    2017-11-14                           i dare you: going bald!?   \n",
       "\n",
       "  publish_time                                               tags    views  \\\n",
       "0   2017-11-13                                    shantell martin   748374   \n",
       "1   2017-11-13  last week tonight trump presidency|\"last week ...  2418783   \n",
       "2   2017-11-12  racist superman|\"rudy\"|\"mancuso\"|\"king\"|\"bach\"...  3191434   \n",
       "3   2017-11-13  rhett and link|\"gmm\"|\"good mythical morning\"|\"...   343168   \n",
       "4   2017-11-12  ryan|\"higa\"|\"higatv\"|\"nigahiga\"|\"i dare you\"|\"...  2095731   \n",
       "\n",
       "    likes  dislikes  comment_count  number_of_tags  \n",
       "0   57527      2966          15954               1  \n",
       "1   97185      6146          12703              36  \n",
       "2  146033      5339           8181              17  \n",
       "3   10172       666           2146              16  \n",
       "4  132235      1989          17518               6  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "trending_date_test = df['trending_date']\n",
    "df['number_of_tags'] = number_of_tags\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,10))\n",
    "hist_numOfTags = plt.hist(df['number_of_tags'],100)\n",
    "fig_views = plt.gcf()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the graph we can see that the distribution is also skewed right. Most of the videos has about 5 to 15 tags, but there are also videos with 80+ tags."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training a regression model\n",
    "With the number of tags, we can then use it as the explanatory variable in a linear regression model and use the number of views as the dependent variable. But first, we will split the whole data set into training and testing sets with a 2:1 ratio, so that we can determine the accuracy of our model later."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "views = np.array(df['views'])\n",
    "numOfTags = np.array(df['number_of_tags'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Coefficients:  [2562.2153975]\n",
      "Mean squared error: 7571400777867.14\n",
      "Variance score: 0.00\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn import datasets, linear_model\n",
    "from sklearn.metrics import mean_squared_error, r2_score\n",
    "\n",
    "# split into train and test sets\n",
    "X_train_tags, X_test_tags, y_train_views, y_test_views = train_test_split(numOfTags, views, test_size=0.33, random_state=42)\n",
    "\n",
    "X_train_tags = X_train_tags.reshape(-1, 1)\n",
    "X_test_tags = X_test_tags.reshape(-1, 1)\n",
    "y_train_views = y_train_views.reshape(-1, 1)\n",
    "y_test_views = y_test_views.reshape(-1, 1)\n",
    "\n",
    "regr = linear_model.LinearRegression()\n",
    "regr.fit(X_train_tags, y_train_views)\n",
    "\n",
    "y_test_views_prediction = regr.predict(X_test_tags)\n",
    "\n",
    "# The coefficients\n",
    "print('Coefficients: ', regr.coef_[0])\n",
    "# The mean squared error\n",
    "print(\"Mean squared error: %.2f\"\n",
    "      % mean_squared_error(y_test_views, y_test_views_prediction))\n",
    "# Explained variance score: 1 is perfect prediction\n",
    "print('Variance score: %.2f' % r2_score(y_test_views, y_test_views_prediction))\n",
    "\n",
    "# Plot outputs\n",
    "plt.figure(figsize=(15,10))\n",
    "plt.scatter(X_test_tags, y_test_views, s= 5, color='red')\n",
    "plt.plot(X_test_tags, y_test_views_prediction, color='blue', linewidth=2)\n",
    "plt.xlabel('Number of tags')\n",
    "plt.ylabel('Number of views')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The result of the regression is not very satisfying. Although the coefficent is indeed positive, we can see that the model's accuracy is extremely low. The red dots in the graph represents the actual (number of tags, number of views) points, while the blue line is the regression line we fit. Looking at the scatter plot, we observe that videos with the same number of tags can have extremely different number of views. This is not surprising because we are not taking the \"quality\" of tags into account here. In other words, some tags might be better than others, and although the number of tags is the same, the effect produced by these tags might differ greatly. Therefore, there is no enough evidence to support our hypothesis for the first question."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analysis II: Linear Regression on Top Tags"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since the number of tags itself doesn't tell the whole story, how about the number of \"useful\" tags? In this report, we decide to measure the \"usefulness\" of a tag by looking at its count (how frequently a tag word is used). We choose this method based on this post: https://whyvideoisgreat.com/optimize-youtube-tags. In the post, the author mentions that we can \"spy on our competitors\" and see what tags they are using. This inspires us that tags that are commonly used might be a sign of popular videos, because the underlying message is that a lot of people are interested in this content. Therefore, we decided to do another regression on the number of TOP tags. But first we need to find out which tags are used more often than others."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Format and clean up tags data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function to remove stop words using the stop_words list\n",
    "def remove_stop_words(list):\n",
    "    filtered_sent = []\n",
    "    for w in list:\n",
    "        if w not in stop_words:\n",
    "             filtered_sent.append(w)\n",
    "    return filtered_sent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "stop_words = stopwords.words('english')\n",
    "tags_words = df['tags'].str.lower().str.cat(sep=' ')\n",
    "tags_words = re.sub('[^A-Za-z]+', ' ', tags_words)\n",
    "tags_words_list = WhitespaceTokenizer().tokenize(tags_words)\n",
    "\n",
    "# Remove stop words\n",
    "filtered = remove_stop_words(tags_words_list)\n",
    "tags_words_series = pd.Series(filtered)\n",
    "\n",
    "# Remove these stop words manually because they were not in the nltk stop words library\n",
    "tags_words_series = tags_words_series.replace(to_replace = 'x', value = '')\n",
    "tags_words_series = tags_words_series.replace(to_replace = 'sp', value = '')\n",
    "tags_words_series = tags_words_series.replace(to_replace = 'b', value = '')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Determine what the most common tag words are"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# use FreqDist to determine the most common tag words\n",
    "tags_count = FreqDist(tags_words_series)\n",
    "tags_count.columns = ['tag','count']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create a line graph to display the most frequently used tag words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2880x1440 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the 100 most frequently used tag words\n",
    "plt.figure(figsize=(40, 20))  \n",
    "tags_count.plot(100, title = 'Most Frequently Used Tag Words', cumulative = False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The graph above provides information about the most freqeuntly used tag words. We decided to observe the 100 words with the highest count. It is interesting to note that the graph begins to normalize and flatten around words that have a count of 2000. From the graph we can see that the top 100 tag words were used at least 1000 times. Thus, we can create a smaller subset of our data that only focuses on tag words that had a count value of 1000 or more. This will make running functions on our data quicker. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The tag word that was used the most frequently is “makeup”. The words “new” and “tutorial” also have a high tag word count. This indicates that making YouTube makeup tutorial videos is a popular right now. We also found that there are a lot of names in our extracted tag words, which also indicates people’s interests in following up celebrities. Additionally, it is interesting that “christmas” is one of the most frequently used tag words. Since Christmas is a seasonal holiday, we expected that it would not be in the top 100 tag words. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create a series with the number of occurrences of each tag\n",
    "series_count = tags_words_series.value_counts()\n",
    "# convert the series into a dataframe\n",
    "df_word_count = pd.DataFrame({'word':series_count.index, 'count':series_count.values})\n",
    "# make a smaller dataframe of only the relevant tag words\n",
    "df_word_count = df_word_count[df_word_count['count'] >= 1000]\n",
    "df_word_count.columns = ['word', 'count']\n",
    "# create a new dataframe with only a column for the most common tag words\n",
    "df_dropped = df_word_count.drop(['count'], axis=1)\n",
    "# convert the df to a list \n",
    "list_100 = df_dropped.values.tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "numOfTopTags = [0 for x in range(len(df.index))]\n",
    "\n",
    "for i in range(len(df.index)):\n",
    "    if i not in df.index:\n",
    "        i += 1\n",
    "        continue\n",
    "    tags_entry = df['tags'][i]\n",
    "    for top_tags in list_100:\n",
    "        if top_tags[0] in tags_entry:\n",
    "            numOfTopTags[i] += 1\n",
    "\n",
    "df.insert(5 , \"number_of_top_tags\" , numOfTopTags)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, with the number of top tags of each video, we can train the regression model just like we did in last section."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Coefficients:  [11538.09911106]\n",
      "Mean squared error: 7566257585382.10\n",
      "Variance score: 0.00\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "array_numOfTopTags = np.array(numOfTopTags)\n",
    "\n",
    "X_train_top_tags, X_test_top_tags, y_train_views, y_test_views = train_test_split(array_numOfTopTags, views, test_size=0.33, random_state=42)\n",
    "\n",
    "X_train_top_tags = X_train_top_tags.reshape(-1, 1)\n",
    "X_test_top_tags = X_test_top_tags.reshape(-1, 1)\n",
    "y_train_views = y_train_views.reshape(-1, 1)\n",
    "y_test_views = y_test_views.reshape(-1, 1)\n",
    "\n",
    "regr = linear_model.LinearRegression()\n",
    "regr.fit(X_train_top_tags, y_train_views)\n",
    "\n",
    "y_test_views_prediction = regr.predict(X_test_top_tags)\n",
    "\n",
    "# The coefficients\n",
    "print('Coefficients: ', regr.coef_[0])\n",
    "# The mean squared error\n",
    "print(\"Mean squared error: %.2f\"\n",
    "      % mean_squared_error(y_test_views, y_test_views_prediction))\n",
    "# Explained variance score: 1 is perfect prediction\n",
    "print('Variance score: %.2f' % r2_score(y_test_views, y_test_views_prediction))\n",
    "\n",
    "# Plot outputs\n",
    "plt.figure(figsize=(15,8))\n",
    "plt.scatter(X_test_top_tags, y_test_views, s = 5, color='red')\n",
    "plt.plot(X_test_top_tags, y_test_views_prediction, color='blue', linewidth=2)\n",
    "plt.xlabel('Number of TOP tags')\n",
    "plt.ylabel('Number of views')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Again, the result did not come out as what we expected. Seems like most of the videos have 0 \"most popular\" (top) tags. Similar to the last regression, the accuracy is low. This also suggests that number of top tags is not a good way to measure the quality of tags. One possible explanation is that we are treating all the popular tags the same way, but in fact their level of popularity is not the same. So we should probably take the acutal count of each tags into account as its weight in the number of top tags. But again, there is no enough evidence to support our second hypothesis."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Analysis III: Sentiment Analysis\n",
    "We chose to preform Sentiment Analysis on our data. Specifically, we looked at the most frequently utilized tag words. \n",
    "\n",
    "We wrote one function to remove irrelevant stop words from our data and another function to preform stemming on the tag words. \n",
    "\n",
    "## Lexicon Analysis\n",
    "\n",
    "compound score - metric that calculates the sum of all the lexicon ratings and normalizes between -1 (most extreme negative) and 1 (most extreme positive).\n",
    "\n",
    "- positive: compound >= 0.05\n",
    "- netural: -0.05 < compound <0.05\n",
    "- negative: <= -0.05"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a function that will output the dataframe that stores sentiment information\n",
    "\n",
    "def get_sentiments(list_in):\n",
    "    list_result = []\n",
    "    for smallList in list_in:\n",
    "        list_result.append(smallList[0])\n",
    "    \n",
    "    analyser = SentimentIntensityAnalyzer()\n",
    "    output = pd.DataFrame()\n",
    "    \n",
    "    for i in range(0, len(list_in)):\n",
    "        word = list_result[i]\n",
    "        ss = analyser.polarity_scores(word)\n",
    "        ss['word'] = word\n",
    "        output = output.append(ss, ignore_index=True)\n",
    "        i = i + 1\n",
    "    return output"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can then use the get_sentiments() function to get the sentiment of the most common tag words."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>compound</th>\n",
       "      <th>neg</th>\n",
       "      <th>neu</th>\n",
       "      <th>pos</th>\n",
       "      <th>word</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>makeup</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>video</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.4404</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>funny</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>new</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>music</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   compound  neg  neu  pos    word\n",
       "0    0.0000  0.0  1.0  0.0  makeup\n",
       "1    0.0000  0.0  1.0  0.0   video\n",
       "2    0.4404  0.0  0.0  1.0   funny\n",
       "3    0.0000  0.0  1.0  0.0     new\n",
       "4    0.0000  0.0  1.0  0.0   music"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# get the sentiment measures\n",
    "sentiment = get_sentiments(list_100)\n",
    "sentiment.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>compound</th>\n",
       "      <th>neg</th>\n",
       "      <th>neu</th>\n",
       "      <th>pos</th>\n",
       "      <th>word</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>69</th>\n",
       "      <td>-0.5574</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>wars</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>145</th>\n",
       "      <td>-0.5994</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>war</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>161</th>\n",
       "      <td>-0.3818</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>battle</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     compound  neg  neu  pos    word\n",
       "69    -0.5574  1.0  0.0  0.0    wars\n",
       "145   -0.5994  1.0  0.0  0.0     war\n",
       "161   -0.3818  1.0  0.0  0.0  battle"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# showing only negative words\n",
    "neg_df = sentiment[sentiment['neg'] == 1]\n",
    "neg_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "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>compound</th>\n",
       "      <th>neg</th>\n",
       "      <th>neu</th>\n",
       "      <th>pos</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>-0.512867</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.115433</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-0.599400</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-0.578400</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-0.557400</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>-0.469600</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>-0.381800</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       compound  neg  neu  pos\n",
       "count  3.000000  3.0  3.0  3.0\n",
       "mean  -0.512867  1.0  0.0  0.0\n",
       "std    0.115433  0.0  0.0  0.0\n",
       "min   -0.599400  1.0  0.0  0.0\n",
       "25%   -0.578400  1.0  0.0  0.0\n",
       "50%   -0.557400  1.0  0.0  0.0\n",
       "75%   -0.469600  1.0  0.0  0.0\n",
       "max   -0.381800  1.0  0.0  0.0"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "neg_df.describe() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looking at our sentiment analysis data, we found that most of the tag words were marked as neutral. The two negative words we observed were \"war\" and \"battle\". This makes sense because a lot of gamers use YouTube to stream their content. The mean value for the negative sentiment words is a bit higher since there are only three negative words, thus we can't rely on the accuracy of this information as much as we can for the positive words since the sample size is so small."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>compound</th>\n",
       "      <th>neg</th>\n",
       "      <th>neu</th>\n",
       "      <th>pos</th>\n",
       "      <th>word</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.4404</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>funny</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.3612</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>comedy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.6369</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>best</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.0772</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>challenge</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>0.5859</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>beauty</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>0.4215</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>entertainment</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>0.5994</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>super</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>0.2023</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>top</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>0.6369</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>love</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>0.5719</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>perfect</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>102</th>\n",
       "      <td>0.4588</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>cute</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110</th>\n",
       "      <td>0.2732</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>humor</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>0.4215</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>marvel</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>0.4404</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>good</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>124</th>\n",
       "      <td>0.5106</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>fun</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>125</th>\n",
       "      <td>0.4939</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>friendly</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>141</th>\n",
       "      <td>0.4588</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>awards</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>151</th>\n",
       "      <td>0.4404</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>easy</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     compound  neg  neu  pos           word\n",
       "2      0.4404  0.0  0.0  1.0          funny\n",
       "9      0.3612  0.0  0.0  1.0         comedy\n",
       "13     0.6369  0.0  0.0  1.0           best\n",
       "15     0.0772  0.0  0.0  1.0      challenge\n",
       "33     0.5859  0.0  0.0  1.0         beauty\n",
       "37     0.4215  0.0  0.0  1.0  entertainment\n",
       "45     0.5994  0.0  0.0  1.0          super\n",
       "66     0.2023  0.0  0.0  1.0            top\n",
       "73     0.6369  0.0  0.0  1.0           love\n",
       "83     0.5719  0.0  0.0  1.0        perfect\n",
       "102    0.4588  0.0  0.0  1.0           cute\n",
       "110    0.2732  0.0  0.0  1.0          humor\n",
       "121    0.4215  0.0  0.0  1.0         marvel\n",
       "123    0.4404  0.0  0.0  1.0           good\n",
       "124    0.5106  0.0  0.0  1.0            fun\n",
       "125    0.4939  0.0  0.0  1.0       friendly\n",
       "141    0.4588  0.0  0.0  1.0         awards\n",
       "151    0.4404  0.0  0.0  1.0           easy"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# showing only positive words\n",
    "pos_df = sentiment[sentiment['pos'] == 1]\n",
    "pos_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "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>compound</th>\n",
       "      <th>neg</th>\n",
       "      <th>neu</th>\n",
       "      <th>pos</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>18.000000</td>\n",
       "      <td>18.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.446178</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.148003</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.077200</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.421500</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.449600</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.556575</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.636900</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        compound   neg   neu   pos\n",
       "count  18.000000  18.0  18.0  18.0\n",
       "mean    0.446178   0.0   0.0   1.0\n",
       "std     0.148003   0.0   0.0   0.0\n",
       "min     0.077200   0.0   0.0   1.0\n",
       "25%     0.421500   0.0   0.0   1.0\n",
       "50%     0.449600   0.0   0.0   1.0\n",
       "75%     0.556575   0.0   0.0   1.0\n",
       "max     0.636900   0.0   0.0   1.0"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos_df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we observe that there are a lot more positive sentiment words in comparison to negative words. This shows that YouTube viewers tend to watch videos for enjoyment. Specifically, the words \"funny\", \"comedy\", and \"humor\" indicate that viewers like watching videos that will make them laugh. The words \"best\", \"super\", \"top\", \"perfect\", and \"good\" are all generic adjectives that can be applicable to a variety of videos. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "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>compound</th>\n",
       "      <th>neg</th>\n",
       "      <th>neu</th>\n",
       "      <th>pos</th>\n",
       "      <th>word</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>makeup</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>video</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>new</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>music</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>show</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   compound  neg  neu  pos    word\n",
       "0       0.0  0.0  1.0  0.0  makeup\n",
       "1       0.0  0.0  1.0  0.0   video\n",
       "3       0.0  0.0  1.0  0.0     new\n",
       "4       0.0  0.0  1.0  0.0   music\n",
       "5       0.0  0.0  1.0  0.0    show"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#showing only neutral words\n",
    "neu_df = sentiment[sentiment['neu'] == 1]\n",
    "# this dataframe is very long so just print the head\n",
    "neu_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "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>compound</th>\n",
       "      <th>neg</th>\n",
       "      <th>neu</th>\n",
       "      <th>pos</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>140.0</td>\n",
       "      <td>140.0</td>\n",
       "      <td>140.0</td>\n",
       "      <td>140.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       compound    neg    neu    pos\n",
       "count     140.0  140.0  140.0  140.0\n",
       "mean        0.0    0.0    1.0    0.0\n",
       "std         0.0    0.0    0.0    0.0\n",
       "min         0.0    0.0    1.0    0.0\n",
       "25%         0.0    0.0    1.0    0.0\n",
       "50%         0.0    0.0    1.0    0.0\n",
       "75%         0.0    0.0    1.0    0.0\n",
       "max         0.0    0.0    1.0    0.0"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "neu_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,10))\n",
    "\n",
    "neg_df['compound'].plot.density(label = 'negative')\n",
    "pos_df['compound'].plot.density(label = 'positive')\n",
    "plt.legend()\n",
    "plt.xlabel('Compound Sentiment Scores')\n",
    "plt.xlim(-1,1);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The positive and negative graphs both have normal distribution with the mean at -0.5 and 0.5 respectively. Since there were more positive words it makes sense that it spans a greater range of compount sentiment words. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Based on our sentiment analysis there are more positive words in the tags in comparison to negative words in which people may assume that people watch videos for entertainment. However, in reality, it is too early to make the conclusion that positive tags will give more views because most of the tag words were actually neutral.\n",
    "\n",
    "Since we used a predetermined list to determine the sentiment of each tag word, we are unsure of how the assignment of each sentiment was given. Therefore, people from different socio-cultural backgrounds may interpret the sentiment of words differently. For instance, the word “family” was assigned a neutral sentiment, but it could also be perceived as having a positive sentiment by some people. Also, because we are separating the tag words from the whole phrase or sentence, this could remove the meaning or the intended sentiment out of the word.\n",
    "\n",
    "In the future, if we could conduct a survey and ask people how they view certain words, as either positive or negative, we can use their opinions to determine the sentiment of the tag words.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Ethics & Privacy\n",
    "### Ethics:\n",
    "Our goal is to see the potential of video tags to influence the views of the videos. Since the data has no underlying assumptions about users, there should be no bias based on different populations. Also, we are only using data of trending Youtube videos in the US within the past few years, and not data from any other other countries. This will prevent us from making generalizations of other countries and places. Last but not least, we can assure that the data we use is approved by Youtube because all information can be accessed on Youtube, except that the dataset collects and integrates into one easy to read file.\n",
    "### Privacy:\n",
    "The data that we used is from a public source and the information that is provided does not have any identifying information so it is ethical for us to use for our analysis. There are names of Youtube channels and authors in the original dataset but this information is public and does not relate to our research question, so we chose to remove it. Also, the topic and hypothesis itself is not related to anything that is personal (tied to any individuals) or potentially vulnerable. Therefore, we don't have any particular concerns regarding privacy. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Conclusion & Discussion"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We hypothesized that more tags in the video will lead to a larger number of views. We also predicted that videos that contain popular tag words will have more views than those videos that don’t have popular tag word. In addition, we predicted that the tags with more positive connotation will have more views than the others. \n",
    "\n",
    "We found while there is a small positive correlation between tags and views, the accuracy of the model is extremely low. Therefore, there is not enough evidence to support the first part of our hypothesis. When determining the “usefulness” of tags, we discovered that using the frequency of top tags is not a good way to measure the quality of tags. In addition to the accuracy of our regression model being low, there is not enough evidence to support our hypothesis. Through sentiment analysis on our dataset, we discovered that most of the tag words were considered ‘neutral’. However, when comparing ‘positive’ and ‘negative’ tag words, there seemed to be more ‘positive’ tag words than ‘negative’ tag words. \n",
    "\n",
    "### Limitation & Posible improvements\n",
    "One of the limitations of our analysis is that our dataset is only based only on trending videos. Thus, any analysis done can only be applied to trending videos and can not be concluded about all of the videos that get posted on YouTube. Additionally, our dataset only includes videos over a short duration of time (several months). If we had data for a longer duration of time this would have given us a more accurate representation. Last, we chose to only utilize the videos that were from the United States. This is a limitation because it is not representative of the entire population, but rather of a subset. Thus, it doesn’t make a good representation of all youtube videos. For improvements in our studies, in the future we could have datasets from a greater time span. Moreover, instead of analyzing only trending videos, we could examine a dataset of all types of YouTube videos."
   ]
  }
 ],
 "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": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}