notebooks/03. more_cleaning.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = pd.read_pickle('../pickle_files/df1.p')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 985,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>date</th>\n",
       "      <th>handle</th>\n",
       "      <th>id</th>\n",
       "      <th>language</th>\n",
       "      <th>tweet</th>\n",
       "      <th>tuples</th>\n",
       "      <th>tweets_clean</th>\n",
       "      <th>positive</th>\n",
       "      <th>negative</th>\n",
       "      <th>neutral</th>\n",
       "      <th>compound</th>\n",
       "      <th>United</th>\n",
       "      <th>Delta</th>\n",
       "      <th>Southwest</th>\n",
       "      <th>American</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-11-04 03:23:36</td>\n",
       "      <td>eirbearrr</td>\n",
       "      <td>59fd32ff291ebbf619c97774</td>\n",
       "      <td>en</td>\n",
       "      <td>@united And the plane is sitting here they won...</td>\n",
       "      <td>(help, one, is, sitting, here, us, the, wont, ...</td>\n",
       "      <td>and the plane is sitting here they wont let us...</td>\n",
       "      <td>0.118</td>\n",
       "      <td>0.096</td>\n",
       "      <td>0.786</td>\n",
       "      <td>0.1280</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-11-04 03:23:34</td>\n",
       "      <td>arlohemphill</td>\n",
       "      <td>59fd32ff291ebbf619c97775</td>\n",
       "      <td>en</td>\n",
       "      <td>@united Sure, DMing now</td>\n",
       "      <td>(dming, sure, now)</td>\n",
       "      <td>sure dming now</td>\n",
       "      <td>0.434</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.566</td>\n",
       "      <td>0.3182</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-11-04 03:22:36</td>\n",
       "      <td>JDR1024</td>\n",
       "      <td>59fd32ff291ebbf619c97776</td>\n",
       "      <td>en</td>\n",
       "      <td>@united United should show the elite status of...</td>\n",
       "      <td>(should, upgrade, of, list, united, show, the,...</td>\n",
       "      <td>united should show the elite status of passeng...</td>\n",
       "      <td>0.205</td>\n",
       "      <td>0.000</td>\n",
       "      <td>0.795</td>\n",
       "      <td>0.4753</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-11-04 03:22:11</td>\n",
       "      <td>loooorenanicole</td>\n",
       "      <td>59fd32ff291ebbf619c97777</td>\n",
       "      <td>en</td>\n",
       "      <td>@united I'm SO DISAPPOINTED in your cust suppo...</td>\n",
       "      <td>(support, in, shame, so, im, on, you, your, cu...</td>\n",
       "      <td>im so disappointed in your cust support shame ...</td>\n",
       "      <td>0.145</td>\n",
       "      <td>0.427</td>\n",
       "      <td>0.429</td>\n",
       "      <td>-0.7398</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-11-04 03:20:36</td>\n",
       "      <td>loooorenanicole</td>\n",
       "      <td>59fd32ff291ebbf619c97779</td>\n",
       "      <td>en</td>\n",
       "      <td>@united INCREDIBLY frustrated w/cust support, ...</td>\n",
       "      <td>(perks, frustrated, wcust, incredibly, explore...</td>\n",
       "      <td>incredibly frustrated wcust support im losing ...</td>\n",
       "      <td>0.214</td>\n",
       "      <td>0.252</td>\n",
       "      <td>0.534</td>\n",
       "      <td>-0.2830</td>\n",
       "      <td>1.0</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": [
       "                 date           handle                        id language  \\\n",
       "0 2017-11-04 03:23:36        eirbearrr  59fd32ff291ebbf619c97774       en   \n",
       "1 2017-11-04 03:23:34     arlohemphill  59fd32ff291ebbf619c97775       en   \n",
       "2 2017-11-04 03:22:36          JDR1024  59fd32ff291ebbf619c97776       en   \n",
       "3 2017-11-04 03:22:11  loooorenanicole  59fd32ff291ebbf619c97777       en   \n",
       "4 2017-11-04 03:20:36  loooorenanicole  59fd32ff291ebbf619c97779       en   \n",
       "\n",
       "                                               tweet  \\\n",
       "0  @united And the plane is sitting here they won...   \n",
       "1                            @united Sure, DMing now   \n",
       "2  @united United should show the elite status of...   \n",
       "3  @united I'm SO DISAPPOINTED in your cust suppo...   \n",
       "4  @united INCREDIBLY frustrated w/cust support, ...   \n",
       "\n",
       "                                              tuples  \\\n",
       "0  (help, one, is, sitting, here, us, the, wont, ...   \n",
       "1                                 (dming, sure, now)   \n",
       "2  (should, upgrade, of, list, united, show, the,...   \n",
       "3  (support, in, shame, so, im, on, you, your, cu...   \n",
       "4  (perks, frustrated, wcust, incredibly, explore...   \n",
       "\n",
       "                                        tweets_clean  positive  negative  \\\n",
       "0  and the plane is sitting here they wont let us...     0.118     0.096   \n",
       "1                                     sure dming now     0.434     0.000   \n",
       "2  united should show the elite status of passeng...     0.205     0.000   \n",
       "3  im so disappointed in your cust support shame ...     0.145     0.427   \n",
       "4  incredibly frustrated wcust support im losing ...     0.214     0.252   \n",
       "\n",
       "   neutral  compound  United  Delta  Southwest  American  \n",
       "0    0.786    0.1280     1.0    0.0        0.0       0.0  \n",
       "1    0.566    0.3182     1.0    0.0        0.0       0.0  \n",
       "2    0.795    0.4753     1.0    0.0        0.0       0.0  \n",
       "3    0.429   -0.7398     1.0    0.0        0.0       0.0  \n",
       "4    0.534   -0.2830     1.0    0.0        1.0       0.0  "
      ]
     },
     "execution_count": 985,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Define positive and negative tweets by their compound score. Above +0.7 is a positive tweet, below -0.4 is a negative tweet. These threshholds gave me a good accuracy of actual positive and negative tweets. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "mask = df['compound'] <= -0.4\n",
    "negative_rows = df.loc[mask] \n",
    "mask = df['compound'] >= 0.7\n",
    "positive_rows = df.loc[mask]\n",
    "mask = (df['compound'] > -0.4) & (df['compound'] < 0.7)\n",
    "neutral_rows = df.loc[mask]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10515\n",
      "6331\n"
     ]
    }
   ],
   "source": [
    "print(len(negative_rows))\n",
    "print(len(positive_rows))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "125340\n",
      "@DogFaceSoldier @Delta And on this end, Delta Flt delayed &amp; missed connecting flt. Bonus is  morning chaos in ATL airport. Not a good sign when food is at gate..\n",
      "-0.6914\n"
     ]
    }
   ],
   "source": [
    "row = np.random.choice(negative_rows.index.values)\n",
    "print(row)\n",
    "print(negative_rows.loc[row, 'tweet'])\n",
    "print(negative_rows.loc[row, 'compound'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "102543\n",
      "@ThePrattPack4 @AmericanAir I really hope tomorrow Wii be better.  Thank you @AmericanAir!\n",
      "0.8221\n"
     ]
    }
   ],
   "source": [
    "row = np.random.choice(positive_rows.index.values)\n",
    "print(row)\n",
    "print(positive_rows.loc[row, 'tweet'])\n",
    "print(positive_rows.loc[row, 'compound'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Sanity check to see if the dates of the tweets are all in the same range."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "mask = df['United'] == 1.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count                   12263\n",
       "unique                  12179\n",
       "top       2017-11-06 00:54:58\n",
       "freq                        3\n",
       "first     2017-10-25 03:17:46\n",
       "last      2017-11-07 03:22:49\n",
       "Name: date, dtype: object"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[mask, 'date'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count                   18044\n",
       "unique                  17852\n",
       "top       2017-10-25 23:15:39\n",
       "freq                        3\n",
       "first     2017-10-25 03:41:37\n",
       "last      2017-11-07 03:22:49\n",
       "Name: date, dtype: object"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mask = df['Delta'] == 1.0\n",
    "df.loc[mask, 'date'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count                   14475\n",
       "unique                  14332\n",
       "top       2017-10-25 20:27:51\n",
       "freq                        4\n",
       "first     2017-10-25 03:54:05\n",
       "last      2017-11-07 03:25:50\n",
       "Name: date, dtype: object"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mask = df['Southwest'] == 1.0\n",
    "df.loc[mask, 'date'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count                   19487\n",
       "unique                  19214\n",
       "top       2017-10-25 15:32:30\n",
       "freq                        4\n",
       "first     2017-10-25 03:27:22\n",
       "last      2017-11-07 03:28:41\n",
       "Name: date, dtype: object"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mask = df['American'] == 1.0\n",
    "df.loc[mask, 'date'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### No weird dates, but there do appear to be some duplicates."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Big cleaning. Prepare the tweets for analysis by removing stop words,  punctuation, numbers, replacing contractions, misspellings, and slang. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "%run ../python_files/cleaning_helper.py"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df1 = df.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "add_leading_and_trailing_space(df1, 'tweets_clean') \n",
    "df1 = filtration_1(df1, 'tweets_clean', 'tweets_clean1')\n",
    "add_leading_and_trailing_space(df1, 'tweets_clean1') \n",
    "contractions(df1, 'tweets_clean1')\n",
    "twitter_words(df1, 'tweets_clean1')\n",
    "airline_words(df1, 'tweets_clean1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "curse_words(df1, 'tweets_clean1')\n",
    "twitter_words(df1, 'tweets_clean1')\n",
    "filtration_2(df1, 'tweets_clean1')\n",
    "strip_leading_and_trailing_spaces(df1, 'tweets_clean1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df1.set_index('date', inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(62689, 15)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(61711, 15)"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1 = df1.drop_duplicates(subset='tweets_clean1', keep=\"first\") \n",
    "df1.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### How many tweets do we have for each day?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df1['time'] = df1.index.time\n",
    "df1['date'] = df1.index.date\n",
    "df1.set_index('date', inplace = True)\n",
    "df1.index.name = 'Date'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Date\n",
       "2017-10-25    5881\n",
       "2017-10-26    5552\n",
       "2017-10-27    5683\n",
       "2017-10-28    4620\n",
       "2017-10-29    4235\n",
       "2017-10-30    4819\n",
       "2017-10-31    3681\n",
       "2017-11-01    3577\n",
       "2017-11-02    4389\n",
       "2017-11-03    4541\n",
       "2017-11-04    3967\n",
       "2017-11-05    4477\n",
       "2017-11-06    5424\n",
       "2017-11-07     865\n",
       "Name: Number of Tweets, dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tweets_per_day_df = df1.groupby('Date').count()['tweet']\n",
    "tweets_per_day_df.rename('Number of Tweets')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7UAAAJXCAYAAABMonMIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x121c2ffd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "labels = tweets_per_day_df.index.values\n",
    "ind = np.arange(len(labels))\n",
    "width = 0.5\n",
    "\n",
    "count = tweets_per_day_df\n",
    "plt.style.use('fivethirtyeight')\n",
    "fig = plt.figure(figsize = (14, 8))\n",
    "ax1 = fig.add_subplot(111)\n",
    "ax1.set_title('Tweets per day', fontsize = 30)\n",
    "ax1.set_xlabel('Date', fontsize = 15)\n",
    "ax1.set_ylabel('Count', fontsize = 15)\n",
    "ax1.bar(ind, count, width)\n",
    "ax1.set_xticks(ind)\n",
    "plt.xticks(rotation = 45)\n",
    "\n",
    "ax1.set_xticklabels(labels);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Make a column for hashtags"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df2 = df1.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df2['hashtags'] = df2['tweets_clean1'].str.findall(r'#.*?(?=\\s|$)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df2['Number of hashtags'] = df2['hashtags'].apply(lambda x: len(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16464"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2['Number of hashtags'].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    61711.000000\n",
       "mean         0.266792\n",
       "std          0.745358\n",
       "min          0.000000\n",
       "25%          0.000000\n",
       "50%          0.000000\n",
       "75%          0.000000\n",
       "max         11.000000\n",
       "Name: Number of hashtags, dtype: float64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2['Number of hashtags'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "mask = df2['hashtags'].astype(str) == '[]'\n",
    "df2.loc[mask, 'hashtags'] = np.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "hashtags = [df2['hashtags'].values.reshape(-1,).tolist()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import itertools\n",
    "unnested = list(itertools.chain(*hashtags))\n",
    "unnested = [x for x in unnested if str(x) != 'nan']\n",
    "unnested = list(itertools.chain(*unnested))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "hashtags_series = pd.Series(unnested)\n",
    "hashtags_list = hashtags_series.value_counts().index.tolist()\n",
    "num_hashtags = hashtags_series.value_counts().tolist()\n",
    "hashtag_df = pd.DataFrame({'Hashtag': hashtags_list, 'Count': num_hashtags})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['#travel', '#avgeek', '#americanairlines', '#ua747farewell',\n",
       "       '#fail', '#customerservice', '#united', '#unitedairlines',\n",
       "       '#airlines', '#southwest'], dtype=object)"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hashtag_df['Hashtag'][:10].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['#travel', '#avgeek', '#americanairlines', '#ua747farewell',\n",
       "       '#fail', '#customerservice', '#united', '#unitedairlines',\n",
       "       '#airlines', '#southwest'], dtype=object)"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "names = hashtag_df['Hashtag'][:10].values\n",
    "names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "hashtag_df = hashtag_df.set_index('Hashtag')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_count_of_hashtags(hashtags):\n",
    "    '''\n",
    "    input: hashtags (list) of hashtags\n",
    "    output: (list) of the count of hashtags    \n",
    "    '''\n",
    "    number = []\n",
    "    for hashtag in hashtags:\n",
    "        number.append(hashtag_df.loc[hashtag, 'Count'])\n",
    "    return number"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABBIAAAIiCAYAAAB18utRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAsIkEAAAAYBMJAAAAwCYSAAAAgC3UBGARGR6f5AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x121ad1518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "labels = list(hashtag_df.index[:10].values)\n",
    "ind = np.arange(len(labels))\n",
    "width = 0.5\n",
    "\n",
    "count = get_count_of_hashtags(names)\n",
    "plt.style.use('fivethirtyeight')\n",
    "fig = plt.figure(figsize = (14, 8))\n",
    "ax1 = fig.add_subplot(111)\n",
    "ax1.set_title('Top Hashtags', fontsize = 30)\n",
    "ax1.set_xlabel('Count', fontsize = 15)\n",
    "ax1.set_ylabel('Hashtag', fontsize = 15)\n",
    "ax1.barh(ind, count, width)\n",
    "ax1.set_yticks(ind)\n",
    "ax1.set_yticklabels(labels)\n",
    "plt.gca().invert_yaxis();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### I didn't do anything to incorporate hashtags into my final analysis. The most common ones were generic. There were a lot of unique hashtags that wouldn't be picked up in the analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df2.to_pickle('../pickle_files/df2.p')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda root]",
   "language": "python",
   "name": "conda-root-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}