EDA_wikipedia_JN_02052017.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Wikipedia Events Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## From João Neves\n",
    "joao.neves\\_at\\_marionete.co.uk"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Wikipedia is the world's largest (free) encyclopedia. Since 2001, Wikipedia has fed the users with the knowledge they needed.\n",
    "\n",
    "To manage this information, someone must fill it in, since Wikipedia is accessible and can be editable by everyone (although there are some protected pages as stated by [Editing](https://en.wikipedia.org/wiki/Help:Editing)).\n",
    "\n",
    "Being a worldwide platform, there are a lot of events involving Wikipedia editions, creations, etc. The data in Websocket [ws://wikimon.hatnote.com:9000](ws://wikimon.hatnote.com:9000) provides a lot of metadata about Wikipedia events. Through this data, we can analyse (for instance) what are the users doing and where from, what kind of entities mess with Wikipedia, how the users describe what are they doing... So let's do that!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "from websocket import create_connection\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "import json\n",
    "import time\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import pickle\n",
    "import re\n",
    "import gensim\n",
    "import random\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Read data from the websocket...! \n",
    "\n",
    "**Note**: the data was previously generated and saved for replication purposes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# wikivents_list = list()\n",
    "\n",
    "# num_minutes = 600\n",
    "\n",
    "# t_end = time.time() + num_minutes*60 #run for n x 60 seconds\n",
    "# ws = create_connection(\"ws://wikimon.hatnote.com:9000\")\n",
    "\n",
    "# while time.time() < t_end:\n",
    "#     wikivents_list.append((get_current_date(),ws.recv())) #interested in time of event\n",
    "    \n",
    "# print(get_current_date())\n",
    "\n",
    "# print(\"Collected: {} events.\".format(len(wikivents_list)))\n",
    "\n",
    "# with open(\"wikimon_data.pkl\",\"wb\") as f:\n",
    "#     pickle.dump(wikivents_list,f)\n",
    "\n",
    "with open(\"wikimon_data.pkl\",\"rb\") as f:\n",
    "    wikivents_list = pickle.load(f)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Create the dataframe"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's parse the *jsons* and create a dataframe from the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "wikivents_list_parsed = map(lambda (date, data): json.loads(data),wikivents_list)\n",
    "wikivents_df = pd.DataFrame(wikivents_list_parsed)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we have a sample of the (final) data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[u'action' u'change_size' u'flags' u'geo_ip' u'hashtags' u'is_anon'\n",
      " u'is_bot' u'is_minor' u'is_new' u'is_unpatrolled' u'mentions' u'ns'\n",
      " u'page_title' u'parent_rev_id' u'rev_id' u'summary' u'url' u'user']\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>action</th>\n",
       "      <th>change_size</th>\n",
       "      <th>is_bot</th>\n",
       "      <th>page_title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8915</th>\n",
       "      <td>edit</td>\n",
       "      <td>-147.0</td>\n",
       "      <td>False</td>\n",
       "      <td>LIBOR market model</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     action  change_size is_bot          page_title\n",
       "8915   edit       -147.0  False  LIBOR market model"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print wikivents_df.columns.values\n",
    "wikivents_df.sample(n=1)[['action', 'change_size', 'is_bot', 'page_title']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the sample above, we see that we have several features related to **Wikipedia events**. We have the page title, the url, the change size, some flags related to the type of event (anonymous, if it was a bot, ...)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Parse geo_ip json\n",
    "*Location* field is a *json*. Here we will expand it and concatenate it to the main Dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "#convert non null locations to dataframe\n",
    "not_null_geoip = wikivents_df[\"geo_ip\"].notnull()\n",
    "location_df = pd.DataFrame(wikivents_df[\"geo_ip\"][not_null_geoip].tolist())\n",
    "\n",
    "#index of wikis that have location\n",
    "indexes = pd.Series(wikivents_df[\"geo_ip\"][not_null_geoip].index.values)\n",
    "location_df = location_df.set_index(indexes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "#join both and drop column\n",
    "wikivents_df = pd.concat([wikivents_df, location_df],axis = 1).drop('geo_ip', 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user</th>\n",
       "      <th>city</th>\n",
       "      <th>country_name</th>\n",
       "      <th>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>region_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>16288</th>\n",
       "      <td>74.219.222.38</td>\n",
       "      <td>Canton</td>\n",
       "      <td>United States</td>\n",
       "      <td>-81.4396</td>\n",
       "      <td>40.8996</td>\n",
       "      <td>Ohio</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53680</th>\n",
       "      <td>2601:644:8103:8FC0:EDB7:B58:18F3:6314</td>\n",
       "      <td>Pleasanton</td>\n",
       "      <td>United States</td>\n",
       "      <td>-121.9374</td>\n",
       "      <td>37.6704</td>\n",
       "      <td>California</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12212</th>\n",
       "      <td>74.211.95.1</td>\n",
       "      <td>Washington</td>\n",
       "      <td>United States</td>\n",
       "      <td>-113.4850</td>\n",
       "      <td>37.2262</td>\n",
       "      <td>Utah</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58457</th>\n",
       "      <td>203.104.11.21</td>\n",
       "      <td>Brisbane</td>\n",
       "      <td>Australia</td>\n",
       "      <td>153.0947</td>\n",
       "      <td>-27.4724</td>\n",
       "      <td>Queensland</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                        user        city   country_name  \\\n",
       "16288                          74.219.222.38      Canton  United States   \n",
       "53680  2601:644:8103:8FC0:EDB7:B58:18F3:6314  Pleasanton  United States   \n",
       "12212                            74.211.95.1  Washington  United States   \n",
       "58457                          203.104.11.21    Brisbane      Australia   \n",
       "\n",
       "       longitude  latitude region_name  \n",
       "16288   -81.4396   40.8996        Ohio  \n",
       "53680  -121.9374   37.6704  California  \n",
       "12212  -113.4850   37.2262        Utah  \n",
       "58457   153.0947  -27.4724  Queensland  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#select some sample data with geo_ip information\n",
    "some_columns = ['user', 'city', 'country_name', 'longitude', 'latitude', 'region_name']\n",
    "# if the 'city' is not null, neither are the others\n",
    "not_null_location = wikivents_df['city'].notnull()\n",
    "wikivents_df[not_null_location].sample(n=4)[some_columns]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The art of Data Science"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we will try extract some knowledge from our data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Action"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "edit         69321\n",
       "hit           3343\n",
       "create        1994\n",
       "delete         532\n",
       "thank          427\n",
       "block          389\n",
       "move           319\n",
       "patrol         205\n",
       "overwrite      182\n",
       "reviewed        90\n",
       "Name: action, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wikivents_df[\"action\"].value_counts()[:10]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Most of the collected events are **edits** in Wikipedia objects that already exist."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Change_size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count     69321.000000\n",
       "mean        217.803609\n",
       "std        4062.765972\n",
       "min     -207155.000000\n",
       "25%           0.000000\n",
       "50%          20.000000\n",
       "75%         156.000000\n",
       "max      256167.000000\n",
       "Name: change_size, dtype: float64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wikivents_df[\"change_size\"].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The changes are almost all around **zero** size."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "null_sizes = wikivents_df[wikivents_df[\"change_size\"].isnull()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>null_actions</th>\n",
       "      <th>action_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>edit</th>\n",
       "      <td>NaN</td>\n",
       "      <td>69321</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      null_actions  action_count\n",
       "edit           NaN         69321"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.concat([null_sizes[\"action\"].rename(\"null_actions\").value_counts(), \n",
    "           wikivents_df[\"action\"].rename(\"action_count\").value_counts()], axis = 1)\n",
    "df[df['null_actions'].isnull()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*change_size* is only filled when the event is an edition and an only edition event has a *change_size*."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Flags"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NaN    51851\n",
       "M      11293\n",
       "B       8264\n",
       "MB      2570\n",
       "N       1652\n",
       "!N      1483\n",
       "NB       106\n",
       "Name: flags, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wikivents_df[\"flags\"].value_counts(dropna = False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Most of the flags are null. The ones that are not null are *minor* (M) and *bot* (B)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>non_null_actions</th>\n",
       "      <th>action_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>edit</th>\n",
       "      <td>25368.0</td>\n",
       "      <td>69321</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      non_null_actions  action_count\n",
       "edit           25368.0         69321"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "not_null_flags_df = wikivents_df[wikivents_df[\"flags\"].notnull()]\n",
    "df = pd.concat([not_null_flags_df[\"action\"].rename(\"non_null_actions\").value_counts(),\n",
    "           wikivents_df[\"action\"].rename(\"action_count\").value_counts()], axis = 1)\n",
    "df[df['non_null_actions'].notnull()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "All the **non-null flags are edits**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--------------------------------------------------\n",
      "True    1652\n",
      "Name: is_new, dtype: int64\n",
      "--------------------------------------------------\n",
      "True    106\n",
      "Name: is_new, dtype: int64\n",
      "--------------------------------------------------\n",
      "True    1483\n",
      "Name: is_new, dtype: int64\n",
      "--------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "print \"-\"*50\n",
    "print wikivents_df[wikivents_df[\"flags\"] == \"N\"][\"is_new\"].value_counts()\n",
    "print \"-\"*50\n",
    "print wikivents_df[wikivents_df[\"flags\"] == \"NB\"][\"is_new\"].value_counts()\n",
    "print \"-\"*50\n",
    "print wikivents_df[wikivents_df[\"flags\"] == \"!N\"][\"is_new\"].value_counts()\n",
    "print \"-\"*50"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The events with flagged *N*, *NB* and *!N* are new events (*is_new* is True)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--------------------------------------------------\n",
      "True    11293\n",
      "Name: is_minor, dtype: int64\n",
      "--------------------------------------------------\n",
      "True    2570\n",
      "Name: is_minor, dtype: int64\n",
      "--------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "print \"-\"*50\n",
    "print wikivents_df[wikivents_df[\"flags\"] == \"M\"][\"is_minor\"].value_counts()\n",
    "print \"-\"*50\n",
    "print wikivents_df[wikivents_df[\"flags\"] == \"MB\"][\"is_minor\"].value_counts()\n",
    "print \"-\"*50"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The events with flagged *M* and *MB* are new events (*is_minor* is True)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Hashtags\n",
    "Hashtags are lists."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    71833\n",
       "1     5376\n",
       "2        9\n",
       "4        1\n",
       "Name: hashtags, dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hashtags_length = wikivents_df[\"hashtags\"].apply(lambda x: len(x))\n",
    "hashtags_length.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "edit       5384\n",
       "byemail       1\n",
       "upload        1\n",
       "Name: action, dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wikivents_df[hashtags_length > 0][\"action\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Most of the events do not have hashtags."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### is_bot\n",
    "\n",
    "From the hashtags, we could see that the hashtag *IABot* exists if a bot edits"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    66279\n",
       "True     10940\n",
       "Name: is_bot, dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wikivents_df[\"is_bot\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True     4849\n",
       "False      28\n",
       "Name: is_bot, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wikivents_df[wikivents_df[\"hashtags\"].apply(lambda x: x == [\"IABot\"])][\"is_bot\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "B    4849\n",
       "Name: flags, dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wikivents_df[wikivents_df[\"hashtags\"].apply(lambda x: x == [\"IABot\"])][\"flags\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*IABot* has flags = *B*. But not the other way around."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>hashtags</th>\n",
       "      <th>is_bot</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>15339</th>\n",
       "      <td>[]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3966</th>\n",
       "      <td>[IABot]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45784</th>\n",
       "      <td>[]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40221</th>\n",
       "      <td>[]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46997</th>\n",
       "      <td>[]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40585</th>\n",
       "      <td>[IABot]</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      hashtags is_bot\n",
       "15339       []   True\n",
       "3966   [IABot]   True\n",
       "45784       []   True\n",
       "40221       []   True\n",
       "46997       []   True\n",
       "40585  [IABot]   True"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wikivents_df_flags_B = wikivents_df[wikivents_df[\"flags\"] == \"B\"]\n",
    "wikivents_df_flags_B.sample(n=6)[['hashtags', 'is_bot']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1    4849\n",
       "0    3415\n",
       "Name: hashtags, dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wikivents_df_flags_B[\"hashtags\"].apply(lambda x: len(x)).value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are *B* flags without hashtags"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    10940\n",
       "Name: is_unpatrolled, dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wikivents_df[wikivents_df[\"is_bot\"] == True][\"is_unpatrolled\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Bots are always patrolled (makes sense), assuming the *patrolled* mean that the event was somehow supervised."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### is_anon"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Feature to flag if the event was made by an anonymous entity."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    63356\n",
       "True     13863\n",
       "Name: is_anon, dtype: int64"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wikivents_df[\"is_anon\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user</th>\n",
       "      <th>is_bot</th>\n",
       "      <th>flags</th>\n",
       "      <th>country_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>13863</td>\n",
       "      <td>13863</td>\n",
       "      <td>33</td>\n",
       "      <td>13852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>6005</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>119</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>86.173.191.67</td>\n",
       "      <td>False</td>\n",
       "      <td>!N</td>\n",
       "      <td>United States</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>70</td>\n",
       "      <td>13863</td>\n",
       "      <td>33</td>\n",
       "      <td>8237</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 user is_bot flags   country_name\n",
       "count           13863  13863    33          13852\n",
       "unique           6005      1     1            119\n",
       "top     86.173.191.67  False    !N  United States\n",
       "freq               70  13863    33           8237"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "some_columns = ['user','is_bot', 'flags','country_name']\n",
    "wikivents_df[wikivents_df[\"is_anon\"]][some_columns].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The anonymous activity does not have a username and are never bots."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### is_unpatrolled"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False    75736\n",
       "True      1483\n",
       "Name: is_unpatrolled, dtype: int64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wikivents_df[\"is_unpatrolled\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "!N    1483\n",
       "Name: flags, dtype: int64"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wikivents_df[wikivents_df[\"is_unpatrolled\"]][\"flags\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Unpatrolled events always have the *!N* flag."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### ns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Main         47258\n",
       "Special       7898\n",
       "User          5685\n",
       "Talk          4824\n",
       "User talk     4193\n",
       "Name: ns, dtype: int64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wikivents_df[\"ns\"].value_counts()[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From this point on, we will just focus on **edition events** since they are the most representative. Let's analyse what kind of things the editors are doing, mainly the **change size**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "wikivents_edits_df = wikivents_df[wikivents_df[\"action\"] == \"edit\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Countries and cities with highest number of events"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x11be47790>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABEcAAAJMCAYAAAAYMch3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11bb2fc50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2) = plt.subplots(ncols=2, figsize=(18,10))\n",
    "\n",
    "n_highest = 10\n",
    "\n",
    "# graphic with the cities\n",
    "df1 = wikivents_edits_df[\"city\"].value_counts()\n",
    "city_order = df1.index[:n_highest]\n",
    "g1 = sns.countplot(y = \"city\", order = city_order, data = wikivents_edits_df, ax=ax1)\n",
    "g1.set_ylabel(\"City\")\n",
    "g1.set_xlabel(\"Number of Events\")\n",
    "\n",
    "# graphic with the countries\n",
    "df2 = wikivents_edits_df[\"country_name\"].value_counts()\n",
    "country_order = df2.index[:n_highest]\n",
    "g2 = sns.countplot(y = \"country_name\", order = country_order,\n",
    "                   data = wikivents_edits_df, ax=ax2)\n",
    "g2.set_ylabel(\"Country\")\n",
    "g2.set_xlabel(\"Number of Events\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We check that London is the city with the most Wikipedia events and the USA are the country with the most Wikipedia events.\n",
    "\n",
    "From the cities with more events, most of them are from the USA."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Countries with the highest average change size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x11c1c5790>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1gAAAGoCAYAAABbkkSYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSZIqMWBJkiRJUiUGLEmSJEmq5P8BJhWykgczFmkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1199d5910>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_highest = 10\n",
    "\n",
    "dict = {\"change_size\":np.mean}\n",
    "\n",
    "# get the data\n",
    "grouped = wikivents_edits_df.groupby([\"country_name\"])\n",
    "df1 = grouped.agg(dict).sort_values([\"change_size\"],\n",
    "                                    ascending = False)[:n_highest]\n",
    "df1[\"country_name\"] = df1.index\n",
    "\n",
    "# plot the data\n",
    "g1 = sns.factorplot(y = \"country_name\", x = \"change_size\",\n",
    "                    data = df1, kind=\"bar\", aspect=2, size=6)\n",
    "g1.set_ylabels(\"Country\")\n",
    "g1.set_xlabels(\"Average Change Size\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Although, when it comes to change size, Mexico is actually the country with the most change size per event in average.\n",
    "\n",
    "So, even if the USA is the country with more events, the change size per event is low. In Mexico, the number of events is low but the change size is high per event."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Cities with the highest average change size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x11c4af5d0>"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1kAAAGoCAYAAAC0UC+mAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noP8HJZAqfwXG41UAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11c9c40d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_highest = 10\n",
    "\n",
    "dict = {\"change_size\":np.mean}\n",
    "\n",
    "# get the data\n",
    "grouped = wikivents_edits_df.groupby([\"city\",\"country_name\"])\n",
    "df = grouped.agg(dict).sort_values([\"change_size\"],\n",
    "                                   ascending = False)[:n_highest]\n",
    "df[\"city_and_country\"] = map(lambda (city, country): city + \" / \" + country,\n",
    "                             df.index)\n",
    "\n",
    "# plot the data\n",
    "g = sns.factorplot(y = \"city_and_country\", x = \"change_size\",\n",
    "                   data = df, kind=\"bar\", aspect=2, size=6)\n",
    "g.set_ylabels(\"City\")\n",
    "g.set_xlabels(\"Average Change Size\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Mexicali in Mexico is the city where change sizes are higher per event although most of the cities are from USA."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Most edited pages (average)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x11cfa9b90>"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAxwAAAHfCAYAAADa5N47AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzs3Xu8plPd+PHPMJgHk+MulMhTfYeiI6JhxpiZhpQhIiKV\nEPUk/SrlmOIhkVJUyCQxIUmIhBmn5CmlcfoWIYkajEhymv37Y63N3e3ehxlzzR57f96v137d930d\n1vVd67r2fq3vtdZ17xHd3d1IkiRJUhMWG+wAJEmSJA1dJhySJEmSGmPCIUmSJKkxJhySJEmSGmPC\nIUmSJKkxJhySJEmSGjNysAOQNDw9/fQz3XPm/Guww1AHK6ywNJ6bRY/nZdHluVl0eW4Wnq6u0SN6\nW+cIh6RBMXLk4oMdgnrhuVk0eV4WXZ6bRZfnZtFgwiFJkiSpMSYckiRJkhozoru7e7BjkDQMzT7x\ndP/4SJK0sGy3daPF+wyHJEmSpEFhwiFJkiSpMSYckiRJkhpjwiFJkiSpMSYckiRJkhpjwiFJkiSp\nMSYckiRJkhpjwiFJkiSpMSYckiRJkhpjwiFJkiSpMSYckiRJkhpjwiFJkiSpMSYckiRJkhpjwqEF\nJiLGR8T0tmVHRsRuC6j82yPipfX9qhHxTERs37L+johYMSLOrZ9nRMSYiDg0IvYa4DHmZdvpEbHk\nfNZlqYi4PyIWr583ioi5EbF+/TwqIu6NiNUi4oS67K66fFpETKnLVo6IU+cnhn7iu7++PnssSZKk\n+WHCoReTXwCb1PdbAufUVyLiVcDszHwoM7ddGMFk5o6Z+eR87vsE8DvgjXXRlsDZ9RVgI+CqzPxr\nZu7dR1FbAj+bnxgkSZIWhpGDHYCGh4joAn5ISXJHAXtl5u8i4uPATkA3MD0zvx4R04CV6s87635b\nAZdSEo4fUTraBwHnRsQIYDxwcT3W/Zm5SlsIW9fRkJWAgzLzp/XzfsAzwNWZuX9LvOOBzwFPAKsD\n3wImAG8AvpaZJ0bEXcCYuu4JYE1gVWC3zLwhIj4GbAssAzwAbNOWoPTU5zfA5sD7gTOAL/TUJyLW\nrO3ytl6adjKwT0RsC3wWeAr4K7AjsBpwYm3vVYEDM/O8iPg9MBNYr7b71sA/ge8ArwPuAJZqOcbe\nEfFpyt+LD2fm7RHxqXqMp4ErM/OzEXF0/XxArduxmXlhL3FLkqRhwhEOLQzdwAbAg8AWwD7AMhGx\nDrADMJbS8Z4aEVH3uTwzN87MOZk5uXbULwc2joiRwKsy8xZgFvBmWhKOXtybmZsD+wIfjYgVKR37\nzTNzLPDyiJjUts8rgPcAHwUOBHap8e/Zofy7M/MdwPHAHhGxGCW5mZiZG1I66+u37XMpsElEvAx4\nLDP/BIyoydk44JI+6kNELAEsnZn/AN4HHF3rcgHwEkoydExmTgL2oLQ7dd2ZmTkOuLfWaRtgVE1s\nPgcs3XKoa2vbHQV8OSLWBd4LbFx/XhMRWwGfBzYDvgdcb7IhSZLAhEML1uP8551xgGXr8p8B1wA/\nAQ4D5gKvB9YALqs/KwGvqftle+GZOYdyB32LWha13LfXsq7vI7bf1Nf7KZ3pVwNdwEURMQNYB/jv\ntn1uysyngIeBO2rSM4cyYtDut/X1HkrHfS7wJHBmRJxCSV6WaNtnVq3vFJ5Lli6hJE+jMvO+PuoD\nJUm7ur7fD5gQETMpScBc4D5gz4j4PrBX2/H/I17gtdT2y8w/1+U9rqyv1wJBSWSuy8ynMrMbuAp4\nXW2r4yhJ5HH9xC5JkoYJEw4tSLcCb4qIVaE8+AxsCtxA6UTfl5mTgS8BR1CSipuBzTJzPDAN+H0t\na24vx7gc+AzPPbdwMbA98Ifaye9Nd9vnOymd6kn12McD1/WzT1/+Y9uIWA+Ympk7AB+n/K6NaN2m\ndtZ/D+zOc/X5GfAJYMYAjrkVZTQDygjGoXXUYgRlxOKLwGmZuQtwRdvx2+t2C+W5ESJiNeDlLes2\nqK+bADcBtwEbRsTIOp1tU+APEbECZZRjP+CkAcQvSZKGARMOLTCZ+Qils3lhRFxNuTN+fGbeDtwI\n7F5HE44G/jczb6SMbFwdEb+m3O2/t73ciPh5y7dBXQq8ldKBJjP/Coymn+lHHWKdDRwLzIyIX1FG\nTf4wbzXu0+3AYxFxTY35PsozFe0uBVav08OgjDKsTd/Tw3q8NjN7Yr4euCAiLgNWoSQiZwNfiYgr\ngUnAyn2U9RPgwdoWx1GeOenxtoi4nDId7TOZOQs4izLKdD1wF3AecArw5cz8BvBQRPzPAOogSZKG\nuBHd3fNyE1eSFozZJ57uHx9JkhaW7bZutPiurtEjelvnCIckSZKkxphwSJIkSWqMCYckSZKkxphw\nSJIkSWqMCYckSZKkxphwSJIkSWqMCYckSZKkxphwSJIkSWqMCYckSZKkxphwSJIkSWqMCYckSZKk\nxphwSJIkSWqMCYckSZKkxozo7u4e7BgkDU/ds2c/OtgxqIOurtF4bhY9npdFl+dm0eW5WXi6ukaP\n6G2dIxySJEmSGmPCIUmSJKkxJhySJEmSGmPCIUmSJKkxJhySJEmSGmPCIUmSJKkxJhySJEmSGmPC\nIUmSJKkxIwc7AEnD023f3HqwQ1AvHhzsANTRUDkvK7339MEOQdJC5giHJEmSpMaYcEiSJElqjAmH\nJEmSpMaYcEiSJElqjAmHJEmSpMaYcEiSJElqjAmHJEmSpMaYcEiSJElqjAmHJEmSpMaYcEiSJElq\njAmHJEmSpMaYcEiSJElqjAmHJEmSpMaMHOwApKEkIpYAvgusCSwFfCkzz4+IVwPTgG7gJmCfzJxb\n93k18OPMXLd+Pg54Yy1yFeDhzHxbyzHGAz8BXp+Z99RlRwK3Zea0iLgL+DMwt+7yUGZuW5ePycx/\n9xL7NODNwEOUvw0PAJ/MzDsHWPe1gIuAXwFfAVbIzCsHsq8kSRq6HOGQFqz3Aw9m5ibAFOAbdfmx\nwIF1+Qhga4CI2AWYDnT1FJCZ+2bmeGAS8A/gIx2O8wRwakSM6CWOyZk5vv5sOw/xf6buMxY4Bjhr\nHvYdC1yYmR8A3gOsMw/7SpKkIcoRDmnBOhs4p74fATxd378FmFnf/wyYDPwYmAOMA+7oUNbHgZ9n\n5qwO6y6n3DDYh+eSmgUqM6+KiKfqCMyBwEr1513AUcDqwKrA+cB3gM8DS0fEQ8BuwJMRcUNmXt9E\nfJIk6cXBhENagDLznwARMZqSeBxYV43IzO76/lFgubr9BXX7/ygnIpYE9gQ26ONwHwWuj4iLO6z7\neUT0TKk6OjMvnPfaAPA3YOX6/vLM/GpErAlcl5m7R8Qo4C+ZeWCd1jUmMw+vU8vuN9mQJEkmHNIC\nFhGrU0YvTsjMM+riuS2bjAYe7qeYicCVmfmP3jbIzAcjYl/ge8A1basn9/asxjxaA/hLzyHr60PA\n+hGxGfAI5VkVSZKkjnyGQ1qAIuJlwM+Bz2bmd1tW/bY+7A2wBXBVP0VNpEy96lNm/pSSCOw2z8H2\nIyImAf/KzJ6Eoydp2o3yIPvOlOc8lu7wLMlc/PsiSZJwhENa0D4PrAAcFBEH1WVbAJ8CTqpTpW7l\nuec8ehPAaQM85r7A5gPc9pqI6JnadUZmHtu2/ssRsT/wDGXq1w4dyrgMOCMiNqI8vP5HYLW2bX4D\nHB0Rt2bmFQOMTZIkDUEjuru7+99Kkhaw2765tX98pGFopfeePtghLHBdXaOZPfvRwQ5DHXhuFp6u\nrtG9fXOmUx4kSZIkNceEQ5IkSVJjTDgkSZIkNcaEQ5IkSVJjTDgkSZIkNcaEQ5IkSVJjTDgkSZIk\nNcaEQ5IkSVJjTDgkSZIkNcaEQ5IkSVJjTDgkSZIkNcaEQ5IkSVJjTDgkSZIkNWbkYAcgaXgas89P\nmD370cEOQx10dY323CyCPC+SXqwc4ZAkSZLUGBMOSZIkSY0x4ZAkSZLUGBMOSZIkSY0x4ZAkSZLU\nGBMOSZIkSY0x4ZAkSZLUGP8Ph6RBcc6pUwY7BEkL2Litzh7sECQtghzhkCRJktQYEw5JkiRJjTHh\nkCRJktQYEw5JkiRJjTHhkCRJktQYEw5JkiRJjTHhkCRJktQYEw5JkiRJjTHhkCRJktQYEw5JkiRJ\njTHhkCRJktQYEw5JkiRJjTHhkCRJktQYE44XuYi4LCI2qO+XjIh/RMSnW9bPiIg3RsT0un5aRExp\nK2P/njLmM4ZREXFXfX9cRLxyHvbdICKejIj1e1k/JiJmDLCs3SLi3X2s77OeEfGxgRynbjsqInYf\n6Pa9lHFoRBz6AvZ/Nob+6t7L/jMiYsz8Hr+fsqdExLQmypYkSS8uIwc7AL1glwKbANfX10uALYGj\nI2IUsAZwY2buCBARzysgM49cUMFk5r7zuMtHgGOAfYDdXuCxp/Wzvr96Hgh8Y4CHWwXYHTh5gNt3\ncj8w9wXs/2wM/dVdkiRpsJhwvPhdChxE6bRvSekAHxURywFvBmZmZncdgXj2bnZEbAh8HdgeOAyY\nTunATgVGAysDh2XmjyJiHHA48AxwB7AnsBTwA2AF4PaWcmcAewH/BE4ERgGrAgdm5nkRsR9we2ae\nHxHLAhOA1wGzImLlzHwgIlatZY+gdMp7yp4FXAmsB9wG/A3YFHii1v2Auv1twGeBJ4G1gOmZeXi9\n4z4d+BNwKvA0ZZRvJ2BXYMWIOIGSvH2orjsEWBvYFlgGeADYph5rnYg4uLbj6cBLKL9TB2bm5RFx\nE/CHGsfx9Rw9BfwL2A74fvvJrDE+AaxZ2223zLwhInYG9q3r/gjs0RbDYrXu367H2gBYEjgkM38S\nEf9LSUgXB47NzLPrIQ+LiJVrubtm5uyIOAYYW9efkZlfi4jXA8fW/VcGPpqZ10bE3bW9bwG+A3wX\neKz+zGmvnyRJGn6cUvXi91tgTESMoHS+ZwK/ACYC44GLO+yzMaXz+K7M/HPbumWAScBk4NiIWAI4\nCdg2M8cB91JGIvYCbsrMTSmd3HZjgGMycxKlc7wPQGYem5nn1212BM7NzH8DPwQ+XJcfAJyZmZsB\n57WUOZrSAd6E0nm+th5/SUrS0moN4D3A24DPtK2bREkqJlISiuUy83Dgoczcu24zJzPHAlcAKwET\nM3NDSkKxPiUBuyUzD6OMjFxaY9keOKWej2WBL9bRpanAWcA4SiK2QmY+lpmPdWi7uzPzHZTEYY+I\nWAn4AjChxvQwJelrjaHHVGDlzNwA2Ax4a0RsAbyq7rsZcEBELF+3PzczJwA/BT4XEVsBr6rtNhbY\nKSLWre37qczcHDgK+GDdf3Vgp8z8JHA0cHBmTgSu7VAvSZI0DJlwvMhl5lzgRmAKcH9mPgH8DHg7\npcP48w67TQaWp9xtbzczM+dm5t8od6hXo9xpP6uOXkymdOZfS+m0k5m/6lDWfcCeEfF9SnKyRIdj\n7Q5sFBEXU5KlPSNisdaygWva9rmhvj5MuatOjXNU23azMvPp2qF/vG3dKXX/i4GPUUY62mWt21zK\nCMWZEXEK8IoOdVmbMvJCZt4LPAK8tLUc4AhKW15GGd3o1PY9fltf76n1Wgu4OTMfrcuv5PkJVo8A\nflljmZOZBwHrAm+p5+/iGv+aLWVBSRCi1uWqzOzOzKeA64B1KInmQRHxvRp/Txs8kJkP1vd9nTdJ\nkjRMmXAMDZcCn6ckGgBXU6ZTLZaZD3XY/lDgq8AJHda9BSAiXkaZIvSX+rN1Zo6n3FW/nNLZ36hu\n+yae3wn/InBaZu5CGSUY0bqy3jVfPDPHZuaUOjpwB7BVa9mU0YRW3R1b4Pn62m5rSqd6c+BsyvQr\n2mKcW+NcD5iamTsAH6f8zoyo63t+f26ljLgQES+nTDN7sLUc4P3AtDpqczNl1Gegsd9JmTq1TP08\njjJVqzWGHrdS2ywilouISyhTnq6o528CZaTljrp9z0P0mwA31f3H1v2XoIyG/ZEybeyQzPwAMIvn\n2qr1GZS+zpskSRqmTDiGhkspncSLADLzScod/Jm97ZCZJ1OeWdipbdUqEXEZcCGwd2Y+A3wCuDAi\nrgX2pnRMvwWsFRFXU6ZLPdFWztnAVyLiSsoUppUBImK/+m1KH+H5zzCcRBlx+BKwTb0jP0/fvDRA\nv6Y8u3A5ZfTl+Lr8log4vW3b24HHIuIaSjvfRxmp+DuwZEQcRRm9mFDreh6wR2a2j5pcD5xc23YC\ncNpAg83MByhTv66IiOsobXliWww9zgfm1PNyCXAcZbrUPyPiKuA3QHfLaMnU2s6TgCMz8wLgzoj4\nJWV045zMvIHyjMrZtYzX1jZo9yngwFrHDQdaP0mSNLSN6O4e6A1jDXURsRswJjP3H+xYNPSdc+oU\n//hIQ8y4rc7uf6MhqKtrNLNnP9r/hlroPDcLT1fX6BG9rXOEQ5IkSVJj/FpcPcv/5SBJkqQFzREO\nSZIkSY0x4ZAkSZLUGBMOSZIkSY0x4ZAkSZLUGBMOSZIkSY0x4ZAkSZLUGBMOSZIkSY0x4ZAkSZLU\nGBMOSZIkSY0x4ZAkSZLUGBMOSZIkSY0ZOdgBSBqetvvgxcye/ehgh6EOurpGe24WQZ4XSS9WjnBI\nkiRJaowJhyRJkqTGmHBIkiRJaowJhyRJkqTGmHBIkiRJaowJhyRJkqTGmHBIkiRJaowJhyRJkqTG\n+I//JA2KQ896x2CHIC1S9tnsnMEOQZIa4QiHJEmSpMaYcEiSJElqjAmHJEmSpMaYcEiSJElqjAmH\nJEmSpMaYcEiSJElqjAmHJEmSpMaYcEiSJElqjAmHJEmSpMaYcEiSJElqjAmHJEmSpMaYcEiSJElq\njAmHJEmSpMYscglHRFwWERvU90tGxD8i4tMt62dExBsjYnpdPy0iprSVsX9PGfMZw6iIuKu+Py4i\nXjkP+24QEU9GxPq9rB8TETMGWNZuEfHuPtb3Wc+I+NhAjlO3HRURuw90+7Z9+2yjiNg0Itabn7Ln\nM55tImK1iFglIk5YWMftI55z52OfbSJitSbiaTnGdRGxZh/r76rXxQv6fZIkScPbyMEOoINLgU2A\n6+vrJcCWwNERMQpYA7gxM3cEiIjnFZCZRy6oYDJz33nc5SPAMcA+wG4v8NjT+lnfXz0PBL4xwMOt\nAuwOnDzA7Vvj6K+NPgRMB34/r2XPp08Ae2XmbcDeC+mYvcrMbedjt08AewF/XcDhzLMF+fskSZKG\nn0U14TiI0mnfktIBPioilgPeDMzMzO46AjGmZ6eI2BD4OrA9cBilg7sKMBUYDawMHJaZP4qIccDh\nwDPAHcCewFLAD4AVgNtbyp1B6fj9EzgRGAWsChyYmedFxH7A7Zl5fkQsC0wAXgfMioiVM/OBiFi1\nlj0CuL+l7FnAlcB6wG3A34BNgSdq3Q+o298GfBZ4ElgLmJ6Zh0fEtFrPPwGnAk9TRq12AnYFVqx3\n+K+ndPoXAw4B1ga2BZYBHgC2qcdaJyIOru14OvASyjVyYGZeHhE3AX8AnuxJ+NraaEfgVcBLKYnh\nJ2v5U4A3R8QtwIbAfrXtr87M/SPiUGBjYFngh8DymfmFiFgKuLG2z561Xt21/l+v9X8CWLOek93q\n6xuB0yLi/cBpmfm2iJgEfAn4N/BgbY83dmpXWkTEdpTkcYl67G3q/t8E3lrPz6uAd9X4jwUWp1xv\nH83MayPi/sxcpbbT74DX17bdvp7zs4DlgKXreViipQ5jM/PJGsu5wNcyc2ZEvJXye7IL5XdkeWA1\n4JuZeWKnY2Xm3RFxeD0f99QYiYhX0OHabmmDafR+nf038Ll6HlYHvkX5HXhDjfVEJEnSsLbITakC\nfguMiYgRlM73TOAXwERgPHBxh302pnT03pWZf25btwwwCZgMHBsRSwAnAdtm5jjgXkpHdS/gpszc\nFPh2h2OMAY7JzEnAHpROKJl5bGaeX7fZETg3M/9N6Th/uC4/ADgzMzcDzmspczRwRmZuQhnNubYe\nf0lK0tJqDeA9wNuAz7Stm0RJKiZSEorlasf5oczsucM/JzPHAlcAKwETM3NDSkKxPiUBuyUzD6OM\njFxaY9keOKWej2WBL7YmGx08kZlbUO7QfzIzf0M5Z5+hJG1fADavsby8JgIAt2bmxsBpwHvr8d4N\nXAC8GtgBGFvbaWo8N7R1d2a+Azge2CMzL6R0tHelJBLUsr7Dc+d8Zq1jf+0K8FrgnTXeW4B31LhW\nyswNKOd49brt64BPZebmwFHABzuUd31mTqQk1u+jdNhXpiQs7wNGttahJ9moTgI+UN9/sH5+NSVR\nmky5xvfr7Vg1SdmUcr53pVx/0Mu13cHzrrO6/BWUNvwopV13AbagJImSJGmYW+QSjsycS7mrPQW4\nPzOfAH4GvJ3S4fx5h90mU+7wPtVh3czMnJuZfwPmUO4CrwqcVe8CT6Z0Ol9L6UyRmb/qUNZ9wJ4R\n8X1KcrJEh2PtDmwUERdTOnZ7RsRirWUD17Ttc0N9fZjSoaXGOaptu1mZ+XRmPgY83rbulLr/xcDH\nKHeg22Wt21xKR/zMiDiF0llsr8valJEXMvNe4BHKqMWz5fTht/X1ng51eDXQBVxU234dSoe7Nb45\ntYyxlETwZMpd+jWAy+rPSsBrBnC8HisDj9S6UOvWk9D11a4Afwe+FxGnUkZalqC0zy9rvLMpI1BQ\nkteDIuJ7wHZ0vkb+I97MvJmS4J4JnEDfv5OXABtExIqUxOtnlBGSqRFxOqWz33rM9rZ5LfDr+vvw\nCDCrrh/ItQ29X2c3ZeZTdd0dNUnqdA1LkqRhaJFLOKpLgc9TOlQAV1OmUy2WmQ912P5Q4KuUDlu7\ntwBExMsoU0v+Un+2zszxlDv7l1M6+xvVbd/E8ztdX6RMz9mFMkowonVlRKwLLJ6ZYzNzSh0duAPY\nqrVsyt3lVt0dW+D5+tpua+Cqemf9bMo0IdpinFvjXA+Ympk7AB+nXAMj6vqe6+FWSoeWiHg5ZZrZ\ng63lzGOcPWXfSen8TqptfzxwXYdyTwL2Bf6rPoeRwM3AZnW/aTz3PEhfx+vxAPCSOrUNYBxlalhv\n+wNQp/F9gTJytTslIRkB3MRz18oKlI48lKloh2TmByid+RHtZbYfr143ozPznZTRi+N7qUNPsng2\nZfrTeZn5DPAp4JeZ+f66rvWY7XW7hZKwLBYRy1ASPujn2m7R23U20GtYkiQNQ4tywjEWuAig3jF9\nmDIVpqPMPJnyzMJObatWiYjLgAuBvWsn7RPAhRFxLeWh4psoc8/XioirKVNKnmgr52zgKxFxJWVq\nSc/89/3qN0l9BPh+2z4nUe4EfwnYpt7V7/Vbp16AXwOHRcTllDvUPZ3WW+qd71a3A49FxDWUdr6P\nMurzd2DJiDgKOAKYUOt6HmWqUqdRk4H6FXAkpc2OBWZGxK8o027+0L5xZs6kjGpMq59vpIxsXB0R\nv6aMbtzbvl+LaylTs1as+3dTzs+5td4TKZ3s/jxCGZH6JXAVJeFYjXItPVCvn1OAf1FGxE4Hzo6I\nqyhJyEC+ZeqPwPja1mcDB7fWoY5mtPou5fmb79bPPwX2iYiZlCTt6frsy/Nk5u8oSfz/UZ7J+Htd\n1fHa7qC360ySJKlXI7q7h+7NyYjYDRiTmfsPdiwaOiJiDPDGzJweEStRRl/WqNP/NECHnvWOofvH\nR5oP+2x2Tp/ru7pGM3v2owspGs0Lz82iy3Oz8HR1je5thsQiO8IhLcruoTyEfR3leYbPmmxIkiR1\ntih+Le4C09//sZDmR33AfOvBjkOSJOnFwBEOSZIkSY0x4ZAkSZLUGBMOSZIkSY0x4ZAkSZLUGBMO\nSZIkSY0x4ZAkSZLUGBMOSZIkSY0x4ZAkSZLUGBMOSZIkSY0x4ZAkSZLUGBMOSZIkSY0ZOdgBSBqe\nDn3vJcye/ehgh6EOurpGe24kSQuMIxySJEmSGmPCIUmSJKkxJhySJEmSGmPCIUmSJKkxJhySJEmS\nGmPCIUmSJKkxJhySJEmSGuP/4ZA0KLb88ZcGOwRpUHxv7CcGOwRJWqgc4ZAkSZLUGBMOSZIkSY0x\n4ZAkSZLUGBMOSZIkSY0x4ZAkSZLUGBMOSZIkSY0x4ZAkSZLUGBMOSZIkSY0x4ZAkSZLUGBMOSZIk\nSY0x4ZAkSZLUGBMOSZIkSY0x4ZAkSZLUmJFNFBoR44G9MnPHlmVHArdl5rQFdIxNgIOBJYBlgFMz\n84QB7DeqxrFmH9ssDvwIOA8Yk5n7t60/NzO37WXfVwJvyMyfDiCWkcClwFLAOzNzTn/7zKuIGAN8\nKzPHR8R0YNfMfLJl/RRgx8zcrZf9RwHvz8yTX0AMRwFbAP+TmTPmt5wmzM912dqmvazfjQ7XTcv6\nZ6+RiDgOODYz/zyPcfdco9sCXwHWpVxHjwH7ZOafetlvPG2/m5IkSU1qJOFoWkSsBXwdmJKZf4uI\n/wKuiIg/ZebFC+AQbweu7W1lb8lGNQEYA/SbcACrAS/JzLfMW3jzZz47masAuwPznXAA21M62I++\ngDKGkmevkczcdz7L6LlGpwCrZeYkgIiYCnwV2HpBBCpJkvRCDSjhiIhlgP8GZgFLZ+Zj83vAiOgC\nfkiZzjWKcrf1dxHxcWAnoBuYnplfj4hpwEr15511v62AXYDTMvNvAJn5eES8A/hnRJwB/CAzL4yI\ntSl3f3cAfgCsANzeEsu6lMRlBPAg8KHM/Ec9xjRgg17qcH9mrhIRewMfAOYC/wd8EtgfWDoirs3M\n81v22RnYF3gC+COwB/At4DUR8e3M3LNl2w2B42ob3QvsXGM5pC5btrbVk8CZwD2U83N9Zn40Ilat\n9R0B3N9S7l2Uju6rgO9S7oY/Bsyp6z9GuWO+DPAAsA1wALBORBwMfA04pZ4PKCMWsyLiVODVwH8B\nX8vM77cc82BKYnVhPUdfAsbW1Wdk5tfaz3PPSE9EvB04BngK+BewHfAeYCowGlgZOCwzfxQR44DD\ngWeAO4Ce9vwW8Jrabgdm5oyIeA9wIDAbWBK4LSKOAO7NzG9GxArAL1oTwT7atLfj9qz/j+sa+CYt\n1wiwH7BXLfN04CWU38sDM/PyiPg9MBNYr5axdds1ugzw1ojYAbgM+AlwUT32XZSRln/3jOQAd1Gu\nuUtqe58o4OakAAAgAElEQVSYmadExIy6fkyt4w61fb4NrA6sCpyfmQdGxKvrsZ8C7gbWrCNo29f6\nPANc3dsIjyRJGl76fYYjIjYHbqR0ZFYB7oqIyfN5vG5Kx/lByhSbfYBlImIdSgdnLLAJMDUiou5z\neWZunJlzMnNynQ60GvAfU0Yy8x+Z+QxwEiUJAPgQpYO8F3BTZm5K6UD1OIky/WQ8pZP2mbp87cy8\nZQD1+SDwsczcCLiV0lE7ktKRbk02VgK+AEzIzLHAw5SO6d7ALa3JRvVtSvKzIXAhsDbwOsrUpvHA\nuZRRA4DXAh+mtOuWEbEKJUk4MzM3o0wLa3c0cHBmTqSO5ETEYpQO6MR63JHA+pTO9C2ZeRjweeCy\nWu4ewIkRMRrYlJKoTKF0Np9V97sfmAxsTkl23kY51zvVpA9aznPL7lOBs4BxwImUhBFKJ3tSLfPY\niFiCci63zcxxlCRtN8rIzAP1vG8NfLNueywwEXgHJZGBMoKza32/EyW5aPW8No2IEb0cl7r+edc1\nJTF73jVCSYAurbFuD5xSy39JPW5P+VvU7dfOzFsy8/+Aj9SybwZ+DWxE35YA3lVj+my9CQBwbb2+\nfkg516sD12XmOyjX1151u6OBI2pbXFPruiLlGt+8XuMvj4hJ/cQhSZKGgYE8NH4EpcP0cGbeR+n8\nHd3PPo9T5pO3WrYu/xmlk/IT4DDK6MDrgTUod2gvo3R8X1P3yw7l303pDD0rIt4QEW8CZlDuyHdR\nOqQ/pXTKrwfIzF9R7sxC6cifUO/ufojSSVqLtmSmDx8E9omImTX+Eb1stxZwc8uUoispCURvVsnM\nW2u8p2TmDZTOZs+oz2aUTiPA7Zn5aE227qOMGj1bX2qHsM3z1mfmXOqISUScAryi5Rg91gU+VNvr\nJGDFWqd9ge9QOqrt573V2sBVmdmdmU8B1wHr1HWdzvMRlOTyMsroRs95m5mZc+sI15y6zarAWTW2\nyZTzsS4lCZtBed5hZN3uocx8MDO7qQlXfebh0Zok7Ayc1l+bAV29HLdHX9d1p7a5ssZyL/AI8NK6\n7rf19R5gVOs1GhHrlV3yfZQbAp+r8bRfi62fr8vMJzPzceAWYM26/PL6ei0QwEPA+hHxA8o0rZ5z\nuzbPTTm8qr6+urbHRbUt1qGMukmSpGFuIAnHYpn57BSSAd75vxV4U52G0vPg8abADcB44L7MnEyZ\nXnMEpbN5M7BZvcM6Dfh9LWtuh/LPAHbvuTMbEctSRgVWrZ3I71OmSv28dmxvod71rUlJT0c6KQ9R\nj6eMblxAmapy4QDqCOXO8l717vObgI1rvO3teiclCVqmfh4H/KGPcv8aEa+p8X42IrahdPA/WB/u\n/ivPdSC7O+z/bH0poxT9rq8d16mZuQPw8VqHEW31uQ34am2v9wKn13P8lszchjLt7cv1YfhObqVO\np6ojDRtTppdB5/P8fmBavZN+M2VUBeAttYyXUUYA/lJ/tq6xHU7pPN9GGR0YTxkZOJsy2rJ8y139\n1vY5CTgI+EtmPtBfm1GmnXU6bo/erutO18itlBEHIuLllNGcB+u69nPceo1OBA6LiMXqtX8z8Fh9\n/29g1Zp8vLFl/zdFxMh6Pa5NmQoGtV0pz4fcTBmteTgzd6ZMbVu6lnVTS1u8rb7eSUmIJtW6Hk9J\nKCVJ0jA3kGc4/hIRWwHdEbE8ZRpUn9+ok5mPRMR+lHn7/6LMkz8+M2+PiDnA9Ij4aD3+YZl5Y0Rc\nBlwdEUtR7iTf215uRPwc2Coz74qIzwDnRsQzlPn8J2fmRXXTaZTOz3r187eA0yLiakon9Im6/KN1\n+UhKp+7DwPvq9j0+EBETWz6Pb3k/C7gqIh6t8f6Kcmf6gIi4ITOn1/Z4ICIOoTzYPpfyHMn+lDvS\nnewJfLduex/leY7T67EeA/5Guavfmy8BP4iIHSkdwXafAr4XEZ+mzNP/d43psYjouXt/Xz3GL4El\no3zT1OGUqT57UDr6h1I68KvU5xGeAb6SmU93CiozL4iI8RHxS8o1cVZm3vDc7LnnuR44udZ5LiXh\nGFePdxmwHLB3Zj4TEZ+gXG+LUc7BrpSRiJPqCNRLgBMy88n6rMolEfEQz42aAPwY+AYl0em3TTNz\nbi/HfWVd39t1PYt6jbSUfwTlnG9HeRZmj8x8upe2Gcdz1+jXKc8p/S4iHqnttEtd92XKVMG7qM/p\nVP+mjDQuDxyamQ/V4+xWf28fq2WsApwRERvx3LNHqwGfrbH+P+AfwFOZOTsijgVmRvkGrbso0+Ek\nSdIwN6K7u9MN8udExEspDwtPBBanTA35nzq9apFU7xCflpmbD3YsWrCin6+cfYFlL015QHvDOsVs\n2KjToPbKzNsGsO3OwK/qDYTdgY0z80Pzeswtf/ylvv/4SEPU98Z+Yr726+oazezZftnfoshzs+jy\n3Cw8XV2je3u0oP8Rjsz8O+Wu/4tCRGxLeXh1r/62lXpExMaUaXlfGG7Jxny4hzJK+S/KqNaHBzke\nSZK0COt1hCMi7qTz8wEAZOZaTQUlaehzhEPDlSMcQ4/nZtHluVl45neEY/yCD0WSJEnScNJrwpGZ\ndwNExI8y8z2t6+qDsD4fIUmSJKlPvSYcEfFj4A2U/03R+n8pRlLmcEuSJElSn/qaUvUBYEXKN1T9\nT8vypylfyypJkiRJferrH/+9OjPvAo6l/Lfknp//pvzDNkmSJEnqU18jHHtR/tHaoR3WdQMTmghI\nkiRJ0tDRV8LxNYDM3GwhxSJJkiRpiOlrStX3F1oUkiRJkoakvhKOXv95hyRJkiQNRF9Tql4ZEd/t\nbWVmfqiBeCRJkiQNIX0lHP8EZi6sQCRJkiQNPX0lHA9m5vcWWiSShpWLtjmQ2bMfHeww1EFX12jP\njSRpgenrGY4nF1oUkiRJkoakXhOOzHzbwgxEkiRJ0tDT1wiHJEmSJL0gJhySJEmSGtPXQ+PPiohl\ngP8GZgFLZ+ZjjUYlSZIkaUjod4QjIjYHbgR+AqwC3BURk5sOTJIkSdKL30CmVB0BjAUezsz7gHHA\n0Y1GJUmSJGlIGEjCsVhm3t/zITNvaTAeSZIkSUPIQJ7h+EtEbAV0R8TywD7An5sNS9JQt9U5Pxjs\nEDQMnTru3YMdgiQNOwMZ4dgT2BlYHfgT8EZgjyaDkiRJkjQ09DvCkZl/B963EGKRJEmSNMT0m3BE\nxB+BxVsWdQOPA7cC/y8z724oNkmSJEkvcgN5huNnlKlU362fdwbWB34KnAJMbCY0SZIkSS92A3mG\nY2xmHpeZj9SfE4H1MvPHwIoNxydJkiTpRWwgCcczEfGOng/1/ZMR8TJgicYikyRJkvSiN5ApVR8E\npkXED4ARwB+B3SjfVPWV5kKTJEmS9GI3kG+pugl4a0SsADyTmY/UVV9sNDJJkiRJL3oD+ZaqscCn\ngWWBERGxOLBGZq7ZcGySJEmSXuQG8gzHycB5lOTkm5QpVT9uMihJkiRJQ8NAEo7HM/NUYAYwB/gI\nMK7JoCRJkiQNDQNJOP4dESsCCbwtM7uBZZoNS5IkSdJQMJCE41jgh5R/9LdrRNwM/LrRqCRJkiQN\nCf0mHJl5NjA5Mx8F3gK8H9il6cAGU0RcFhEb1PdLRsQ/IuLTLetnRMQbI2J6XT8tIqa0lbF/Txnz\nGcOoiLirvj8uIl45D/tuEBFPRsT6vawfExEzBljWbhHx7j7W91nPiPjYQI5Ttx0VEbsPdPteyjg0\nIg59Afs/G0N/de9l/xkRMWZ+j99P2VMiYlrL5x9GxNIRcf98lHVuh2V7zU/bRcQ2EbHavO4nSZKG\nh16/pap2IL4BvAa4OiI+l5kPA79dWMENokuBTYDr6+slwJbA0RExClgDuDEzdwSIiOcVkJlHLqhg\nMnPfedzlI8AxwD6U/5nyQo49rZ/1/dXzQMp1NBCrALtTvqhgft0PzH0B+z8bQ391H0z1OlwsM//V\n6frrT2ZuuwDD+QSwF/DXBVimJEkaIvr6WtxTgd8A3wF2AL5K+SeAw8GlwEGUTvuWlA7wURGxHPBm\nYGZmdtcRiGfvZkfEhsDXge2Bw4DplA7sVGA0sDJwWGb+KCLGAYcDzwB3AHsCSwE/AFYAbm8pdwal\nQ/dP4ERgFLAqcGBmnhcR+wG3Z+b5EbEsMAF4HTArIlbOzAciYtVa9ghKp7yn7FnAlcB6wG3A34BN\ngSdq3Q+o298GfBZ4ElgLmJ6Zh9c77tOBP1GumacpI2c7AbsCK0bECZTk7UN13SHA2sC2lOeBHgC2\nqcdaJyIOru14OvASynV6YGZeHhE3AX+ocRxfz9FTwL+A7YDvt5/MGuMTwJq13XbLzBsiYmdg37ru\nj5R/Ztkaw2K17t+ux9oAWBI4JDN/EhH/S0lIFweOraOBAIdFxMq13F0zc3ZEHAOMrevPyMyvRcTr\nKVMWF6dcGx/NzGsj4u7a3rdQfv++CzxWf+bUMiYAVwywnh8GPlqPc35mHhIR92fmKvVrr79Wy30a\nuK6W9fF6Drvruf56p/Lr6xuB0yJiYj1nywFLAwdk5s/bz4ckSRpe+ppS9fLM/HxmXkzpiM339KAX\nod8CYyJiBKXzPRP4BTARGA9c3GGfjSmdx3dl5p/b1i0DTAImA8dGxBLAScC2mTkOuJfSedsLuCkz\nN6V0ctuNAY7JzEmUc7IPQGYem5nn1212BM7NzH9Tnr35cF1+AHBmZm5G+ZrjHqMpHeBNKJ3na+vx\nl6QkLa3WAN4DvA34TNu6SZSkYiIloVguMw8HHsrMves2czJzLKWjvBIwMTM3pCQU61MSsFsy8zDK\nyMilNZbtgVPq+VgW+GIdXZoKnEX51rQTgRUy87HMfKxD292dme+gJA57RMRKwBeACTWmhylJX2sM\nPaYCK2fmBsBmlH+EuQXwqrrvZsABEbF83f7czJxAee7pcxGxFfCq2m5jgZ0iYt3avp/KzM2Bo3gu\noV8d2CkzPwkcDRycmROBa1tieidwwQDq+VJgf8q5fTOwVE1Ke5wIvK+WfydARKxDuckwtu43NZ4b\nRvmP8jPzQuB3lORydUri9C7gfQzg//xIkqShr6+E48meN5n5VOvnoS4z5wI3AlOA+zPzCeBnwNsp\nnbBOd20nA8tT7ra3m5mZczPzb5Q7yatR7gyfVUcvJlM686+ldNrJzF91KOs+YM+I+D4lOVmiw7F2\nBzaKiIspydKeEbFYa9nANW373FBfH6bcVafGOaptu1mZ+XTt0D/etu6Uuv/FwMcod8vbZa3bXMr1\ndGZEnAK8okNd1qaMvJCZ9wKPAC9tLQc4gtKWl1FGNzq1fY+eqYD31HqtBdxcn02iHqs9weoRwC9r\nLHMy8yBgXeAt9fxdXONfs6UsKAlC1LpclZnd9XfpOmAdSqJ5UER8r8bf0wYPZOaD9X1v5+2VHRLb\n3up5U2Y+Xo+/f2b+s2X7l2XmH9rKfz3lerys/qxEmVrZqfxnZebNlET5TOAEBvalFJIkaYiblw5B\nd2NRLJouBT5PSTQArqbcIV4sMx/qsP2hlGlnJ3RY9xaAiHgZZYrQX+rP1pk5nnJX/XJKZ3+juu2b\neH4n/IvAaZm5C2WUYETrynrXfPHMHJuZU+rowB3AVq1lU0YTWg303Pa13daUTvXmwNmU6Ve0xTi3\nxrkeMDUzdwA+TrkOR9T1PdfkrZS760TEyynTzB5sLYfyBQbT6qjNzZRRn4HGfidl6lTPVzyPo0zV\nao2hx63UNouI5SLiEsqUpyvq+ZtAGWm5o27fMxq4CXBT3X9s3X8JymjYHynTxg7JzA8As3iurVqf\nQXneeavtN2uA9byDMlq3VN33nNqePe6NiLVby6ckdDcDm9X6TQN+30v5PfEuVq+/0Zn5TuADlFEQ\nSZI0zPWVcLwuIv7U89Py+c76eai7lNJJvAggM5+k3MGf2dsOmXky5ZmFndpWrRIRlwEXAntn5jOU\nB20vjIhrgb0pHdNvAWtFxNWU6VJPtJVzNvCViLiSMoVpZYCI2K9+m9JHeP4zDCdRRhy+BGxT78jP\n0zcvDdCvKc8uXE4ZfenpbN4SEae3bXs78FhEXENp5/soIxV/B5aMiKMooxcTal3Po0zfaR81uR44\nubbtBOC0gQabmQ9Qpn5dERHXUdryxLYYepwPzKnn5RLgOMp0qX9GxFWUZ526W0ZLptZ2ngQcmZkX\nAHdGxC8poxvnZOYNlOcdzq5lvLa2QbtPAQfWOm5Yl21F5+lUneo5mzJda2Y9/g11xKjHnpTnLy6j\njGqQmTdSRjaujohfU0Y37qV311La/kFgfD1nZwMHDyRGSZI0tI3o7u580zoi1uhrx8y8u5GIhpiI\n2A0Yk5n7D3Ys0qJkq3N+MNxGTbUIOHVcE/dbFo6urtHMnv1o/xtqofPcLLo8NwtPV9foEb2t6/Wh\nThMKSZIkSS+U3yLTsEX5fzlIkiRJTfNbZCRJkiQ1xoRDkiRJUmNMOCRJkiQ1xoRDkiRJUmNMOCRJ\nkiQ1xoRDkiRJUmNMOCRJkiQ1xoRDkiRJUmNMOCRJkiQ1xoRDkiRJUmNGDnYAkoanC7bbmdmzHx3s\nMNRBV9doz40kaYFxhEOSJElSY0w4JEmSJDXGhEOSJElSY0w4JEmSJDXGhEOSJElSY0w4JEmSJDXG\nhEOSJElSY0w4JEmSJDXGf/wnaVBs86OrBzsEDSPf2fQNgx2CJA1bjnBIkiRJaowJhyRJkqTGmHBI\nkiRJaowJhyRJkqTGmHBIkiRJaowJhyRJkqTGmHBIkiRJaowJhyRJkqTGmHBIkiRJaowJhyRJkqTG\nmHBIkiRJaowJhyRJkqTGmHBIkiRJaowJh4aViBgfEdPblh0ZEbstwGOcEBG/HcB20yNiyV7WHRoR\nf4iIGRFxdURcFBHL91HWqIjYveXzmyPisPmrwbyJiOsiYs2FcSxJkvTiY8IhLUARsTQwFrg1Isb3\ntW1m7piZT/axybGZOT4zxwK/A3bvY9tV2tZvBVwwsKglSZKaM3KwA5AWEV0RcTklCR8F7JWZv4uI\njwM7Ad3A9Mz8ekRMA1aqP+8EfghsVZOH9wKXAT8DPgbMiIj1gK9l5mYAEXEBcBDwY2AMsDpwMrAk\n8C9gxw7xrQDcVvffGdgXeAL4I7AHcACwTkQcnJmHAW8FvhgRewMfAOYC/5eZ/1PjH1GPuyywa2be\n1ktdVwe+A/wX8DiwR2beExGHA1OAe4CVa1zX1PU3R8QWwLsyc+/5ORmSJGnocIRDKv4OPAhsAewD\nLBMR6wA7UEYsNgGmRkTU7S/PzI0zc05mTm4Zqdidkjz8AnhTRLw8M38PjIqINSJiVWDlzGydcvUV\n4H8zcyPga8Cb6vL96pSq3wPvAi6PiJWALwAT6sjHw8CewOHALZl5WES8DPh7ZnYDHwQ+Vsu+NSJ6\nbjLckZkTgEOBL/dR168AX8/M8fX9kRHxVmBTYH1gV2B0LfNkSnID8KH6WZIkDXMmHBpuHgeWalu2\nbF1+DfAT4DDKiMDrgTUoIxaXUUY0XlP3yfaCI2Ltus8xwEWUkYK96upTKJ3zXYBT23cFfgmQmedn\n5s/r8p4pVesBBwPTgLWAmzPz0brNlcDr2sp7Zz0+lIRjn4iYWesyoi6/vL5eW4/fW13XBT4fETNq\nDC8DXgv8OjPnZuYjwKxa1lnAuyPipcArMvOG9jaSJEnDjwmHhptbKSMPq0J52Jpyt3554L7MnAx8\nCTiCklTcDGxW7/BPA35fy5nboezdgQMyc0pmTgEmAB+qD4ZPpzxXsQ1wRoeY1q/x7FynNrW7hzLl\n6k7K1Kll6vJxwB9qPD2/z5OAnqTlI5TpYeMoIycb1+Vvqa9vr3Xsra63AZ+ty/YEzgZuATaIiMVq\nHOsAZOZjwBWUUZrTO9RBkiQNQyYcGlbqHfn9gAsj4mrKCMHxwI+A3eud/KMpU5xupNztvzoifk25\n439ve5kR8fOaVLyP8jxHz7H+DNwIbJeZ/6zvb20ZnejxaeBz9dg7Az+oy3umVP2CMo3qU5n5AHAI\ncEVEXEd5fuJEypSwJSPiq/z/9u48TK6qzv/4uyFAjETG/KYFQSQIwxdwFAFZDRB2iUhYBkVQJ4qy\nyojLAI4gCDrCz4GRRQmrEVFgZHFBCCAYCCiLQGT1i2yCYKDZJAKGreePc/qh6OmNkNuV0O/X8/RT\nVbfuPed761LhfuqcWwWL1/6gjD7MrNenPApcV5dvU5cdUNvtb1+/DBxaR0jOAG7JzFmUa1RuoASp\nR1v25RRgcss+SJKkEa6ju7u73TVIGkb1ovGzM3N6A22vA+yXmZ8cbN0dzrvaf3w0bE7eeI12l/C6\ndXaOpaur9+cVWhB4bBZcHpvh09k5tqO/5/yWKknzRUR8Dtid8k1dkiRJgIFDGnEyc0pD7Z4AnNBE\n25IkaeHlNRySJEmSGmPgkCRJktQYA4ckSZKkxhg4JEmSJDXGwCFJkiSpMQYOSZIkSY0xcEiSJElq\njIFDkiRJUmMMHJIkSZIaY+CQJEmS1BgDhyRJkqTGjGp3AZJGpgt2mkBX15x2l6E+dHaO9dhIkuYb\nRzgkSZIkNcbAIUmSJKkxBg5JkiRJjTFwSJIkSWqMgUOSJElSYwwckiRJkhpj4JAkSZLUGAOHJEmS\npMb4w3+S2uK7FzzS7hLUr2fbXcCQfWTCmHaXIEkahCMckiRJkhpj4JAkSZLUGAOHJEmSpMYYOCRJ\nkiQ1xsAhSZIkqTEGDkmSJEmNMXBIkiRJaoyBQ5IkSVJjDBySJEmSGmPgkCRJktQYA4ckSZKkxhg4\nJEmSJDXGwCFJkiSpMaPaXYDeGCJiIrBXZu7SsuxI4A+ZOW0+tH83sGFmPhoRbwf+DOySmT+pz98D\nrAOcmpk7RsQMYC9gF2B2Zk4dQh+HvYZ1zwY+mZnPz8O+LAH8CVguM1+KiA2Aa4D1MvOGiBgN9OzP\nwZm5T0TcD6wKTAXOzszpEfGPwLeBfYATgWWBMcBsYM/MfPy11tZS417AMsC02t/689qWJEka2Rzh\n0MLiV8BG9f4k4Nx6S0SsCHRl5hOZueNwFJOZu8xL2KjbzgVmAe+riyYBP6m3ABsAMzPz4czcZ4Cm\nJgEXA5+iBKWtMnMCcDXwtXmpTZIkaX5zhEONi4hO4BxKwB1NGQmZFRH7AbsC3ZRP0Y+LiGnA/6t/\nH6rbbQtcRgkc51FOtA8Bzo+IDmAiML32NTszl+lVwuSI2Lm2eUhm/qI+/iLwEnB1Zh7UUu9E4CvA\nXGB5yqjCZsAawLGZeWKvEYe5wHjg7cCUzLwpIj4H7Ai8GXgM2KFXQOnZnxuBzYGPAz8Gvt6zPxEx\nnoFHF7YC9gW2AD4TEdcAVwLHAx11X/5PHfU1n0QZDVkJOCozp0XEBOBY4EngReDa2k9nRPwcWBq4\nMDOPqLWdTvk3pBv4N+AJ4ApgY2C1ui+bZuaL/dQvSZJGAEc41LRuYF3gcWAbygnymyNideCjwATK\niff2ERF1mysyc8PMfLJ+av885UR2w4gYBayYmXcAtwJr0RI4+vFQZm4O7A/sHRHjKCfDm9cRgeUi\nYste27wD2AnYGzgY+EStf88+2v9TZm5NOdHfIyIWoYSbLTJzPcpJ+Tq9trkM2Cgilgaeycx7gY4a\nzjYBLhlgf4iIxYAxmfnXzDwP+AawO3AfcDmw2iB1LJWZ2wLbAT1h60TgY5m5RW2nx5J1/zcEtomI\nNYD/ooSvjYHPA6dl5oPAAcAPgP+ubRk2JEka4Qwcml+eA5botWzJuvxiyjUKPwMOB14G/hlYgXJy\nfDnlxPif6nbZu/HM7PnUfZvaFrXdD9S2rh+gthvr7WzKp/orA53ARfVaj9Upn/S3ui0zXwCeAu6p\noedJyghNbzfX2weB0Zn5MvA8cFZEnEYJL4v12ubWur8f5JWwdAklPI3OzL8MsD9QQtrVAPUakMsz\nc3vgbZTrLqYNUses1prr/aUz8656v+c1Bvh9DTYvUV7nVSgjGFcBZOYsykgQwE9rP1dm5p8H2QdJ\nkjQCGDg0v9wJrFkv6KZe+LwxcBPlJPovmbkV5ZP4/6SEitspU24mUk6Sb6ltvdxPH1dQPkG/uD6e\nDuwM3FVPrvvT3evxfZQT7S1r38fzyvSh/rYZyKvWjYj3Attn5keB/Sjvs47WdTKzm7K/n+GV/bmY\nMlowYwh9bgtcWO9/rG5HDQW3AHMHqaOv/XsoIlar91tHZFaLiCXr6NJ6lON2J/Wamoh4HyXMAXwJ\nuBR4f0R4obkkSTJwaP7IzKcp10T8MiKupnz6fXxm3g38nnKNwQzKtyp9KzN/TxnZuDoifkf5tP+h\n3u1GxKURsXh9eBnwfuDXtc+HgbEMMv2oj1q7gGOAKyPiOsqoyV0Db/Wa3A08U6+puAz4C+UbpHq7\nDFi+Tg+DMnqwGgNPD+uxSstoxFeBlSJiVu3zSMr0qqHW0WNP4IyIuJwy+tTjCcq1NL8Bzq31fhnY\nLyKuokzF2j0i3k+5PuTA2v/pEbHUEPZFkiS9gXV0d7+WD3Ilaf747gWP+I+PXrePTBjT7hKGTWfn\nWLq65rS7DPXBY7Pg8tgMn87OsR39PecIhyRJkqTGGDgkSZIkNcbAIUmSJKkxBg5JkiRJjTFwSJIk\nSWqMgUOSJElSYwwckiRJkhpj4JAkSZLUGAOHJEmSpMYYOCRJkiQ1xsAhSZIkqTEGDkmSJEmNGdXu\nAiSNTPvusDRdXXPaXYb60Nk51mMjSZpvHOGQJEmS1BgDhyRJkqTGGDgkSZIkNcbAIUmSJKkxBg5J\nkiRJjTFwSJIkSWqMgUOSJElSYwwckiRJkhrjD/9JaoubT3203SWoH3/muXaXMGTvmPymdpcgSRqE\nIxySJEmSGmPgkCRJktQYA4ckSZKkxhg4JEmSJDXGwCFJkiSpMQYOSZIkSY0xcEiSJElqjIFDkiRJ\nUmMMHJIkSZIaY+CQJEmS1BgDhyRJkqTGGDgkSZIkNcbAIUmSJKkxo9pdgDQSRMQqwKHA4sA1mfmd\nNsYphNUAABGTSURBVJckSZI0LAwcEhARE4GfAf+cmQ/WZUcCf8jMaa+3/cy8KyKeB9YBDhhiTW8G\nrgf2ycwr67IfAjMz8+QBtjsQ2B/4SGbOHGJf04C1gCeAbmBRYG/gGeDszFy/n+2+BlycmTcMpR9J\nkjTyOKVKesVc4PsR0TG/G46ItwIzgMnASkPZJjOfAT4OnB4R4yJiX6B7kLAxCniKEmze/hrLPCAz\nJ2bmpsC3gCOGUOPhhg1JkjQQRzikV1xBCeH7Aie0PhERhwA7AF3AGOAQYCIwOzOnRsSqwFRgD+DM\nzFy3bncOcDTwJuCzwKeBeyLiSmA3YFJtbyXgqN6jKZl5c0QcD5wLvAXYuLY7BVg1Mw+KiNGUkZjx\nwAeAjwIfA5aMiFnA88BZwIO1n+szc+9BXotxwN/q/c6I+CklwNySmZ+tIyJnZ+b0iPggsEtmTomI\n3SkjI4sCP8/MQwfpR5IkvcE5wiG92t7AFyJi5Z4FEbEGsA1l1GB7Bhg5yMy7gOciYvWIGAesCNwA\nnALsmJmbAA8BU+omS2XmtsB2wEH9NDsNWBc4PzOfHaT+dwMfz8yJwPnAznX5KsDutZ1JEbFMH9v+\n/4iYERGXAx8EDqzL3wJ8CtgA2Dwi3tZXx3X5QcBGlOlZS0TEkoPUK0mS3uAc4ZBaZObjEbE/8APg\nmrp4NcqowEuUMPG7PjZtnYZ1CiVQPACcCXRSQsr/RASU0Y7LgLuBWXWbB4HR/ZR1GvB1YO+IuDAz\nbxmg74eA4yLib8ByLftwd2bOAYiIv/TT1wGZOb11QUSMB+7NzCfr40cpIzJ99f8u4LbMfK4+7i9A\nSZKkEcQRDqmXzPwFkLwyCnE7sE5ELBIRSwBr1uV/55XRjrVamjgX2IoyBetM4DHgz8DkOvLwTcr0\nLSgXaPcrIr4AvJiZ3wb2BM6KiDED9H0K8KnMnAI8zCthYMB+BtHXtn31fw+wan2NiIhzI2K519Gv\nJEl6AzBwSH3bH3gOIDNvBS4CrgUuAF6of+dQpifNoOWkPzP/DlwFPJqZT2Tmy8DngV9GxG+AfYDb\nBisgItYH9qJcF0JmXgZcTLm+ZDowPiKuBj4CPF03OxOYGRHXAGOBZef9JRjQqZSpZ7+ijKSQmV3A\nUcCVEfFb4KbMfKih/iVJ0kKio7v79XzwKb3x1WsT/iUzv1c/vb8d2CwzHxhgm+8C52XmFf2tM9Ld\nfOqj/uOj1+0dk9/U7hKGTWfnWLq65rS7DPXBY7Pg8tgMn87Osf1+y6fXcEiDe4wypeoGyvSiUwcJ\nG5cCjxk2JEmSDBzSoOqUqE+9hvW3arAcSZKkhYrXcEiSJElqjIFDkiRJUmMMHJIkSZIaY+CQJEmS\n1BgDhyRJkqTGGDgkSZIkNcbAIUmSJKkxBg5JkiRJjTFwSJIkSWqMgUOSJElSY0a1uwBJI9Oan3kb\nXV1z2l2G+tDZOdZjI0mabxzhkCRJktQYA4ckSZKkxhg4JEmSJDXGwCFJkiSpMQYOSZIkSY0xcEiS\nJElqjIFDkiRJUmMMHJIkSZIa4w//SWqLR75zY7tLUD8eaajdRXZbpaGWJUkLMkc4JEmSJDXGwCFJ\nkiSpMQYOSZIkSY0xcEiSJElqjIFDkiRJUmMMHJIkSZIaY+CQJEmS1BgDhyRJkqTGGDgkSZIkNcbA\nIUmSJKkxBg5JkiRJjTFwSJIkSWqMgUOSJElSYwwcakRETIyIs/tYfnZELN7PNuMiYtfX0MdZEXFD\nRKw6yHoHRcS6vZaNjoj7h9rXwiYi7o+I0b2WTYmIIyNimYj43gDbzhjsNZUkSRqqUe0uQCNLZu4y\nwNPvBbYDfjzE5rbIzM4h9HnkENsbETJzNrBPu+uQJEkjg4FDw6qOKqwKTAIOBF4AHgZ2Ab4KrBER\ne2TmyS3bbAl8A/g78DjwaeA/gaUi4meZObmutxhwJ7BGZj4TEV8GXgLWAM4GrgZ+BLwVuLul/fcA\nxwEdPe1n5l8j4mhgQl3tx5l5bK992RY4HPgr8CRwC3AEcBKwPPB24OeZeXBETKv7ugKwRK3nw8A7\ngcmZeU9EfAvYCFgUOCYzf9Krv3uB64CVgNuAzwBfA2Zn5tQ6KjE1MyfWTU6KiPHAI8C/trQzHjg7\nM9ePiG8Cm1L+LTgvM4+qqx0aEUsDbwY+lpn39lVfRMwAHgXGAfsCpwMvUkZPd83MB5EkSSOaU6rU\nLh8Dvp2ZE4ALgbcA3wSu6BU2OoCTgR0zcxPgSuDgzNwHeKInbABk5gvAecBOddGuwBktfe4F3JaZ\nG1NCQY9TgH3rifpFwAE1TKwIrE8JHbvWYNJT16KUkLJNZm4KPFefWh64NjO3Btatffa4PzO3ooSi\nFTNzUq33wxGxTV02gRIAvhoR/9DrNXsHcEhmrgssCWzfz2vb48T6mt0PfLafdXajvE4bAU+1LP9l\nZm4GXAz8yyD1nZWZWwBbANfX20OBpQapT5IkjQAGDrXLF4HNIuJKYEPg5X7W+0fg6cx8qD6+Cnj3\nAO2eCnyyXrORmfl4y3OrUE6IyczrKCMOAKsB36uf1n8aWK4um5mZ3TXIXAus3tJWZ63rkfp4Zr19\nAlgnIn4E/DdlNKPHTfX2KeCOev9JYDTwHmDtWsN0YDFgfK99eyAze0ZmfgNEr+c7Wu4/n5nXDrBu\nj92AI4FLgNaAc2O9nQ2MGaS+rLen1X2bDnyOMtIhSZJGOAOH2mUP4LD6CXwHsAMldPT+b/Ix4C0R\n8fb6eBPgrv4azcw/1vb+nTJy0eoOYAOAiFiTctIM5YT5k3WE4wDKiMud1OlUdarWhsAfW9p6FBgb\nET3XkKxfb6cAT2XmbsDRwJg6SgPQ3V/dwB+AX9caNgP+B7in1zrLRcQy9f4HgNsp08x6Xpu1WtZd\nPCLeV+9vRJmC9SoRsQSwM2W0aVNgSkSs0E+tA9XXExYnU0La5sBPKFPmJEnSCGfgUJO2iojftfyt\n0vLc9cCFEXE5sAzlJP8e4D0RsX/PSpnZTZkOdH5EXEOZrnPEIP2eBqwJ/LrX8qnAuyLiasr1BnPr\n8r2BM+ryI4FbMvNC4L6I+C1ldOPczOwZoSAzX6Z8in9RRPyKci3GC8DlwAcj4irgREpIWXbQVwp+\nAfwtImZSRhe6M3NOr3XmAidExHWU615+AZwDTKojD2v1Wne/WsfbKCM/r5KZcykjMtfW1+pS4IHX\nUd/vgMMj4grKVLLjh7DfkiTpDa6ju3ugD10l9ScivkK5eHpuRJwJXJqZZwy23evob3ZmLjP4mguH\nR75zo//4jDCL7LbK4CupX52dY+nq6p3ztSDw2Cy4PDbDp7NzbEd/z/ktVdK8mwNcGxHPUi7MPqe9\n5UiSJC14DBzSPMrME4AThrG/N8zohiRJGjm8hkOSJElSYwwckiRJkhpj4JAkSZLUGAOHJEmSpMYY\nOCRJkiQ1xsAhSZIkqTEGDkmSJEmNMXBIkiRJaoyBQ5IkSVJjDBySJEmSGjOq3QVIGpmW3n9turrm\ntLsM9aGzc6zHRpI03zjCIUmSJKkxBg5JkiRJjTFwSJIkSWqMgUOSJElSYzq6u7vbXYMkSZKkNyhH\nOCRJkiQ1xsAhSZIkqTEGDkmSJEmNMXBIkiRJaoyBQ5IkSVJjDBySJEmSGjOq3QVIGlkiYhHge8Aa\nwFzgM5l5d3urGtki4ibg6frwPuAk4FjgReDSzPx6u2obiSJiPeCozJwYESsD04Bu4DZg38x8OSIO\nBT5EOUb7Z+b1bSt4BOl1bNYELgT+WJ8+MTPP8dgMr4hYDDgdGA8sAXwDuAPfNwsUA4ek4bY9MDoz\nN4iI9YGjgcltrmnEiojRQEdmTmxZNgvYCbgX+GVErJmZN7epxBElIg4APgE8UxcdAxycmTMiYiow\nOSL+BGwCrAcsD5wHrNOOekeSPo7N2sAxmXl0yzpr4bEZbh8HHs/MT0TEOGBW/fN9swBxSpWk4TYB\nmA6QmdcC729vOSPeGsCYiLg0Iq6IiI2BJTLznszsBi4BtmhviSPKPcCOLY/XBq6s9y+mHIsJlJGn\n7sx8ABgVEZ3DW+aI1Nex+VBEXBURp0XEWDw27fAT4JB6v4MyeuH7ZgFj4JA03N4C/LXl8UsR4Whr\n+zwL/BewNbAX8P26rMccYKk21DUiZeZ5wAstizpq8INXjkXv95DHaBj0cWyuB/49MzemjAYeisdm\n2GXm3zJzTg185wIH4/tmgWPgkDTcngbGtjxeJDNfbFcx4i7gzPqp312U/yGPa3l+LPBUWyoTwMst\n93uORe/3kMeoPS7IzBt77gNr4rFpi4hYHvg18MPM/DG+bxY4Bg5Jw+0aYBJAvYbj1vaWM+J9mnId\nDRGxLDAGeCYiVoqIDsrIx8w21jfS3RwRE+v9bSjH4hpg64hYJCLeSQntj7WrwBHskohYt97fHLgR\nj82wi4ilgUuBAzPz9LrY980CxmkMkobbBcCWEfEbynzbT7W5npHuNGBaRFxN+UaXT1M+HfwRsChl\nzvN1baxvpPsScEpELA7cCZybmS9FxEzgt5QPDvdtZ4Ej2N7A8RHxAjAb2CMzn/bYDLv/AN4KHBIR\nPddyfB44zvfNgqOju7t78LUkSZIkaR44pUqSJElSYwwckiRJkhpj4JAkSZLUGAOHJEmSpMYYOCRJ\nkiQ1xq/FlSRpARcR44H7gK0y87KW5fcDEzPz/tfZ/nxpZ5A+3kn5vYRnal9z6vIpwDHAA7022bP3\nVzJHxDRgRm3n1MycFBEfBv4pM4+JiL0AMnNqU/sh6bUzcEiStHB4gfKbHO/pOVlfyEwEbsrMXft4\n7ueZOWWoDWXmw9QfEAXWbllu0JAWQAYOSZIWDg8Dl1F+GX6P1ifqryoflpkT6+NplJGAGcBPgXuB\n9wC/q8umUH4sbYfMvLM2c1hErAH8nTK6cEv9FeeTgOUpPwj5lcz8VUQcBqwPvBM4ITO/11LLKsDJ\nwDjKaMa/UcLSN4AlI2JqZu41lB2uv3Z/NLBt3f9FgRl1xGcGJXTsVdf9E7ACQGYeFhHb1j4Xqfu/\nZ2Y+UkdzfghsDbwZ+GRm3hgRXwT+te7n9Zm551BqlDQ4r+GQJGnh8SVg64jY8jVs817gCCCAdYDx\nmbkBcBavDi5/zMw167o/qMuOBU7PzLWB7YCTImJsfW50Zq7eGjaqM4HjMvO9wBeAcym/9vw1ykhG\nX2Fju4iY1fLXM5VqJ2BN4N3AzsDKrRtl5h3AVGBqZn6/Z3lEvI0SlLavdVwDnNCy6eOZuW7d9j8i\nYhTwFeD9lBGTlyNiuT7qlDQPDBySJC0kMvNp4LOUqVVjB1u/mp2ZN2fmy8Cfgcvr8j9RRjl6nFr7\nuAhYISL+AdgCODwiZgEXA4sBK9X1X3V9BUBELAmsnJnn17auBZ6ghJ2B/Dwz39fyt15dPhE4PzNf\nyMwu4KIh7vO6lFGK++vjk4HNW56fXm9vA8Zl5ovAb4AbgEOB72bmQ0PsS9IgDBySJC1EMvNSXpla\n1aMb6Gh5vFjL/ed7NfFiP033Xv48ZQrTZj1BgDKN6tb6/HN9tLFIrzqoj+d1Cnc3rz5X6a/2vuoY\nqIa/t7TfU+/2wN718fSI2OS1lSqpPwYOSZIWPl+iXIOwbH38GPCuiBgdEeOAjeahzd0AImIH4A+Z\n+SxwBbBPXb46cAswpr8G6gjMPRGxY91mfWAZykjCvPgVsHNELBERbwU+2Mc6L/J/A811wPr1Wg8o\nU8d+3V8nEdFJmfZ1a2Z+jfItWO+dx5ol9eJF45IkLWQy8+mI+CxwSX18e0T8ErgduB+YOQ/NrlKn\nTs2hXDwNsB9wckTcQvnk/xOZOSdiwBlSHwemRsTXgbnAjpn5/CDbbFf7bnVMZp4REetQAsts4I4+\ntr0K+EFEPNKzoF4cvgdwQUQsTpk+tnt/nWdmV0ScBNwQEc9SvqJ32kAFSxq6ju7u7nbXIEmSJOkN\nyilVkiRJkhpj4JAkSZLUGAOHJEmSpMYYOCRJkiQ1xsAhSZIkqTEGDkmSJEmNMXBIkiRJaoyBQ5Ik\nSVJj/heNuspopsMlegAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11c4af510>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(9,8))\n",
    "\n",
    "n_highest = 10\n",
    "\n",
    "df = wikivents_edits_df[\"page_title\"].value_counts()\n",
    "page_order = df.index[:n_highest]\n",
    "g = sns.countplot(y = \"page_title\", data = wikivents_edits_df, order = page_order)\n",
    "g.set_ylabel(\"Page Title\")\n",
    "g.set_xlabel(\"Number of Editions\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The most edited page is a user page with more than 200 editions. There are just three main pages in the top 10: *2017 NFL Draft*, *Nguyēn Xuân Phüc*, and *List of video game publishers*."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Most active users (number of events)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAoIAAAHfCAYAAADECLmhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11fc79a10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(9,8))\n",
    "n_highest = 10\n",
    "\n",
    "# indexes of the users sorted by number of events\n",
    "user_order = wikivents_edits_df[\"user\"].value_counts().index[:n_highest]\n",
    "\n",
    "# plot with the \"is bot\"'s\n",
    "## select the respective data\n",
    "is_bot = wikivents_edits_df[\"is_bot\"]\n",
    "is_bot_sorted = wikivents_edits_df[is_bot][\"user\"].value_counts().loc[user_order]\n",
    "## plot the data\n",
    "g1 = sns.barplot(y = is_bot_sorted.index, x = is_bot_sorted, color=\"red\")\n",
    "g1_legend = plt.Rectangle((0,0),1,1,fc=\"red\", edgecolor = 'none')\n",
    "\n",
    "# plot with the non \"is bot\"'s\n",
    "## select the respective data\n",
    "is_not_bot = wikivents_edits_df[\"is_bot\"]==False\n",
    "is_not_bot_sorted = wikivents_edits_df[is_not_bot][\"user\"].value_counts().loc[user_order]\n",
    "## plot the data\n",
    "g2 = sns.barplot(y = is_not_bot_sorted.index, x = is_not_bot_sorted, color=\"blue\")\n",
    "g2_legend = plt.Rectangle((0,0),1,1,fc='blue',  edgecolor = 'none')\n",
    "\n",
    "g2.set_ylabel(\"User\")\n",
    "g2.set_xlabel(\"Number of Events\")\n",
    "l = plt.legend([g1_legend, g2_legend], ['Is a bot?','Is not a bot?'],\n",
    "               loc=5, ncol = 1, prop={'size':11})\n",
    "sns.despine(left=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The most active user is a bot which looks like some kind of *archive bot*."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Most active users (average change per event)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x11c995810>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1gAAAGoCAYAAABbkkSYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\niCZJghUREREREdEk/wWE+yRSblLpqQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11cf48710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_highest = 10\n",
    "\n",
    "dict = {\"change_size\":np.mean}\n",
    "\n",
    "# get the data\n",
    "grouped = wikivents_edits_df.groupby([\"user\"])\n",
    "average_change_size_by_user = grouped.agg(dict).sort_values([\"change_size\"]\n",
    "                                                            ,ascending = False)[:n_highest]\n",
    "average_change_size_by_user[\"user\"] = average_change_size_by_user.index\n",
    "\n",
    "# plot the data\n",
    "g = sns.factorplot(y = \"user\", x = \"change_size\",\n",
    "                   data = average_change_size_by_user,\n",
    "                   kind=\"bar\", aspect=2, size=6)\n",
    "g.set_ylabels(\"User\")\n",
    "g.set_xlabels(\"Average Change Size\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Non-anonymous users are responsible for higher chance sizes per event in average."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Let's play a little..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's grab the summary of the editions and check what a Word2Vec model trained with these texts would be..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, let's make some basic NLP processing in the summary text."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# pattern to remove everything that is not words and whitespaces\n",
    "pattern = re.compile(r\"[^a-zA-Z\\s]*\")\n",
    "\n",
    "# select only non null summaries\n",
    "df = wikivents_edits_df[wikivents_edits_df[\"summary\"].notnull()][\"summary\"]\n",
    "sentences = map(lambda record: pattern.sub(lambda match: \"\", record), df.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Our training data has 57493 sentences.\n"
     ]
    }
   ],
   "source": [
    "training_sentences = []\n",
    "\n",
    "# lower case and split the sentences\n",
    "for sentence in sentences:\n",
    "    training_sentences.append(sentence.lower().split())\n",
    "    \n",
    "print \"Our training data has {} sentences.\".format(len(training_sentences))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Train the model. We must be aware that such a complex model needs a lot of data for training and some iterations which is not the case. This is just an ilustration of an interesting approach."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# we are just training to generate features of size two (which is very low)\n",
    "# but this is just because of time restrictions\n",
    "\n",
    "# model = gensim.models.Word2Vec(training, size=2, window=6, min_count=5)\n",
    "# model.save(\"model_example.model\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**note**: the model was previously trained (with the parameters below) and locally saved to reuse."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "model = gensim.models.Word2Vec.load(\"model_example.model\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def collect_word_vecs(model):\n",
    "    for word in model.wv.vocab.keys():\n",
    "        yield model.wv[word]\n",
    "        \n",
    "def collect_vocabulary_words(model):\n",
    "    for word in model.wv.vocab.keys():\n",
    "        yield word\n",
    "\n",
    "word_vecs = list(collect_word_vecs(model))\n",
    "\n",
    "words = list(collect_vocabulary_words(model))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(u'restoration', 0.9999997019767761),\n",
       " (u'coast', 0.9999996423721313),\n",
       " (u'issue', 0.9999996423721313),\n",
       " (u'festivals', 0.9999996423721313),\n",
       " (u'shift', 0.9999995827674866),\n",
       " (u'make', 0.9999994039535522),\n",
       " (u'navy', 0.9999994039535522),\n",
       " (u'wpnpov', 0.9999992847442627),\n",
       " (u'wikilove', 0.9999985694885254),\n",
       " (u'advert', 0.9999985098838806)]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.wv.most_similar(\"portugal\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For some reason, the models thinks that *portugal* is near *festivals*, *restoration* and *coast* which is somehow interesting because Portugal is in the west edge of Europe, is famous for its restoration and for its music festivals."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's see the word representation of some of these words."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA5sAAAJTCAYAAACGpwJ9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKUmSJEmKuv8FcjBOixdO68cAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11fcfac50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# some words taken from the dictionary (variable 'words')\n",
    "some_words = [\"talkthe\",\"specialcontributionsjax\",\"categoryyouth\",\n",
    "              \"categoryosmania\",\"don\",\"slang\",\"pt\",\"host\",\"blog\",\"autumn\",\n",
    "             \"categorylists\",\"sourcing\",\"hate\",\"provider\",\"saying\",\n",
    "             \"mode\",\"usermiftermifter\",\"repair\",\"international\",\"bio\"]\n",
    "indexes = [i for i in range(len(words)) if words[i] in some_words]\n",
    "\n",
    "# get the data\n",
    "x = map(lambda (x,y): x,word_vecs)\n",
    "y = map(lambda (x,y): y,word_vecs)\n",
    "x_sample = [x[i] for i in indexes]\n",
    "y_sample = [y[i] for i in indexes]\n",
    "\n",
    "# plot the data\n",
    "plt.figure(figsize=(15,10))\n",
    "plt.scatter(x_sample,y_sample,marker=\"o\")\n",
    "plt.title(\"Word space representations (sample)\", size = 20)\n",
    "\n",
    "# annotate the points with the word they are representing\n",
    "for i, w in enumerate([words[i] for i in indexes]):\n",
    "    plt.annotate(\n",
    "        w.decode('utf-8', 'ignore'),\n",
    "        xy = (x_sample[i], y_sample[i]), xytext = (3, 3),\n",
    "        textcoords = 'offset points', ha = 'left', va = 'top', fontsize = 15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We conclude that the model capture some word semantics and syntax. For instance, we have some notion about the proximity of the words according to our vocabulary. For instance, *categoryyouth* is near *categoryosmania*, probably because they have *category* in common. On the other side *bio* is far from *slang*. "
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda env:ipykernel_py27]",
   "language": "python",
   "name": "conda-env-ipykernel_py27-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}