Sentiment analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "import pandas as pd \n",
    "import numpy as np \n",
    "import matplotlib.pyplot as plt \n",
    "import string\n",
    "import nltk\n",
    "import warnings \n",
    "warnings.filterwarnings(\"ignore\", category=DeprecationWarning)\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df = pd.read_csv('train.csv')\n",
    "test_df = pd.read_csv('test.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "combi = train_df.append(test_df, ignore_index=True, sort=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>label</th>\n",
       "      <th>tweet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>#fingerprint #Pregnancy Test https://goo.gl/h1...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Finally a transparant silicon case ^^ Thanks t...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>We love this! Would you go? #talk #makememorie...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>I'm wired I know I'm George I was made that wa...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>What amazing service! Apple won't even talk to...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9868</td>\n",
       "      <td>9869</td>\n",
       "      <td>NaN</td>\n",
       "      <td>#SamsungGalaxyNote7 Explodes, Burns 6-Year-Old...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9869</td>\n",
       "      <td>9870</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Now Available - Hoodie. Check it out here - ht...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9870</td>\n",
       "      <td>9871</td>\n",
       "      <td>NaN</td>\n",
       "      <td>There goes a crack right across the screen. If...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9871</td>\n",
       "      <td>9872</td>\n",
       "      <td>NaN</td>\n",
       "      <td>@codeofinterest as i said #Adobe big time we m...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9872</td>\n",
       "      <td>9873</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Finally I got it .. thanx my father .. #Samsun...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>9873 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        id  label                                              tweet\n",
       "0        1    0.0  #fingerprint #Pregnancy Test https://goo.gl/h1...\n",
       "1        2    0.0  Finally a transparant silicon case ^^ Thanks t...\n",
       "2        3    0.0  We love this! Would you go? #talk #makememorie...\n",
       "3        4    0.0  I'm wired I know I'm George I was made that wa...\n",
       "4        5    1.0  What amazing service! Apple won't even talk to...\n",
       "...    ...    ...                                                ...\n",
       "9868  9869    NaN  #SamsungGalaxyNote7 Explodes, Burns 6-Year-Old...\n",
       "9869  9870    NaN  Now Available - Hoodie. Check it out here - ht...\n",
       "9870  9871    NaN  There goes a crack right across the screen. If...\n",
       "9871  9872    NaN  @codeofinterest as i said #Adobe big time we m...\n",
       "9872  9873    NaN  Finally I got it .. thanx my father .. #Samsun...\n",
       "\n",
       "[9873 rows x 3 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combi"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>label</th>\n",
       "      <th>tweet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>#fingerprint #Pregnancy Test https://goo.gl/h1...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>Finally a transparant silicon case ^^ Thanks t...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>We love this! Would you go? #talk #makememorie...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>I'm wired I know I'm George I was made that wa...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>What amazing service! Apple won't even talk to...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7915</td>\n",
       "      <td>7916</td>\n",
       "      <td>0</td>\n",
       "      <td>Live out loud #lol #liveoutloud #selfie #smile...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7916</td>\n",
       "      <td>7917</td>\n",
       "      <td>0</td>\n",
       "      <td>We would like to wish you an amazing day! Make...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7917</td>\n",
       "      <td>7918</td>\n",
       "      <td>0</td>\n",
       "      <td>Helping my lovely 90 year old neighbor with he...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7918</td>\n",
       "      <td>7919</td>\n",
       "      <td>0</td>\n",
       "      <td>Finally got my #smart #pocket #wifi stay conne...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7919</td>\n",
       "      <td>7920</td>\n",
       "      <td>0</td>\n",
       "      <td>Apple Barcelona!!! #Apple #Store #BCN #Barcelo...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7920 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        id  label                                              tweet\n",
       "0        1      0  #fingerprint #Pregnancy Test https://goo.gl/h1...\n",
       "1        2      0  Finally a transparant silicon case ^^ Thanks t...\n",
       "2        3      0  We love this! Would you go? #talk #makememorie...\n",
       "3        4      0  I'm wired I know I'm George I was made that wa...\n",
       "4        5      1  What amazing service! Apple won't even talk to...\n",
       "...    ...    ...                                                ...\n",
       "7915  7916      0  Live out loud #lol #liveoutloud #selfie #smile...\n",
       "7916  7917      0  We would like to wish you an amazing day! Make...\n",
       "7917  7918      0  Helping my lovely 90 year old neighbor with he...\n",
       "7918  7919      0  Finally got my #smart #pocket #wifi stay conne...\n",
       "7919  7920      0  Apple Barcelona!!! #Apple #Store #BCN #Barcelo...\n",
       "\n",
       "[7920 rows x 3 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def remove_pattern(input_txt, pattern):\n",
    "    r = re.findall(pattern, input_txt)\n",
    "    for i in r:\n",
    "        input_txt = re.sub(i, '', input_txt)\n",
    "        \n",
    "    return input_txt    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# remove twitter handles (@user)\n",
    "combi['tidy_tweet'] = np.vectorize(remove_pattern)(combi['tweet'], \"@[\\w]*\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# remove special characters, numbers, punctuations\n",
    "combi['tidy_tweet'] = combi['tidy_tweet'].str.replace(\"[^a-zA-Z#]\", \" \")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "combi['tidy_tweet'] = combi['tidy_tweet'].apply(lambda x: ' '.join([w for w in x.split() if len(w)>3]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>label</th>\n",
       "      <th>tweet</th>\n",
       "      <th>tidy_tweet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>#fingerprint #Pregnancy Test https://goo.gl/h1...</td>\n",
       "      <td>#fingerprint #Pregnancy Test https MfQV #andro...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.0</td>\n",
       "      <td>Finally a transparant silicon case ^^ Thanks t...</td>\n",
       "      <td>Finally transparant silicon case Thanks uncle ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>0.0</td>\n",
       "      <td>We love this! Would you go? #talk #makememorie...</td>\n",
       "      <td>love this Would #talk #makememories #unplug #r...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>I'm wired I know I'm George I was made that wa...</td>\n",
       "      <td>wired know George made that #iphone #cute #dav...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>What amazing service! Apple won't even talk to...</td>\n",
       "      <td>What amazing service Apple even talk about que...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id  label                                              tweet  \\\n",
       "0   1    0.0  #fingerprint #Pregnancy Test https://goo.gl/h1...   \n",
       "1   2    0.0  Finally a transparant silicon case ^^ Thanks t...   \n",
       "2   3    0.0  We love this! Would you go? #talk #makememorie...   \n",
       "3   4    0.0  I'm wired I know I'm George I was made that wa...   \n",
       "4   5    1.0  What amazing service! Apple won't even talk to...   \n",
       "\n",
       "                                          tidy_tweet  \n",
       "0  #fingerprint #Pregnancy Test https MfQV #andro...  \n",
       "1  Finally transparant silicon case Thanks uncle ...  \n",
       "2  love this Would #talk #makememories #unplug #r...  \n",
       "3  wired know George made that #iphone #cute #dav...  \n",
       "4  What amazing service Apple even talk about que...  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combi.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    [#fingerprint, #Pregnancy, Test, https, MfQV, ...\n",
       "1    [Finally, transparant, silicon, case, Thanks, ...\n",
       "2    [love, this, Would, #talk, #makememories, #unp...\n",
       "3    [wired, know, George, made, that, #iphone, #cu...\n",
       "4    [What, amazing, service, Apple, even, talk, ab...\n",
       "Name: tidy_tweet, dtype: object"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tokenized_tweet = combi['tidy_tweet'].apply(lambda x: x.split())\n",
    "tokenized_tweet.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    [#fingerprint, #pregnanc, test, http, mfqv, #a...\n",
       "1    [final, transpar, silicon, case, thank, uncl, ...\n",
       "2    [love, thi, would, #talk, #makememori, #unplug...\n",
       "3    [wire, know, georg, made, that, #iphon, #cute,...\n",
       "4    [what, amaz, servic, appl, even, talk, about, ...\n",
       "Name: tidy_tweet, dtype: object"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from nltk.stem.porter import *\n",
    "stemmer = PorterStemmer()\n",
    "\n",
    "tokenized_tweet = tokenized_tweet.apply(lambda x: [stemmer.stem(i) for i in x]) # stemming\n",
    "tokenized_tweet.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       [#fingerprint, #pregnanc, test, http, mfqv, #a...\n",
       "1       [final, transpar, silicon, case, thank, uncl, ...\n",
       "2       [love, thi, would, #talk, #makememori, #unplug...\n",
       "3       [wire, know, georg, made, that, #iphon, #cute,...\n",
       "4       [what, amaz, servic, appl, even, talk, about, ...\n",
       "                              ...                        \n",
       "9868    [#samsunggalaxynot, explod, burn, year, thank,...\n",
       "9869    [avail, hoodi, check, here, http, zetasuppli, ...\n",
       "9870    [there, goe, crack, right, across, screen, cou...\n",
       "9871             [said, #adob, time, well, includ, #appl]\n",
       "9872    [final, thanx, father, #samsung, #galaxi, #gif...\n",
       "Name: tidy_tweet, Length: 9873, dtype: object"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tokenized_tweet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       #fingerprint #pregnanc test http mfqv #android...\n",
       "1       final transpar silicon case thank uncl #yay #s...\n",
       "2       love thi would #talk #makememori #unplug #rela...\n",
       "3       wire know georg made that #iphon #cute #davent...\n",
       "4       what amaz servic appl even talk about question...\n",
       "                              ...                        \n",
       "9868    #samsunggalaxynot explod burn year thank rush ...\n",
       "9869    avail hoodi check here http zetasuppli product...\n",
       "9870    there goe crack right across screen could actu...\n",
       "9871                    said #adob time well includ #appl\n",
       "9872    final thanx father #samsung #galaxi #gift #fat...\n",
       "Name: tidy_tweet, Length: 9873, dtype: object"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "for i in range(len(tokenized_tweet)):\n",
    "    tokenized_tweet[i] = ' '.join(tokenized_tweet[i])\n",
    "\n",
    "tokenized_tweet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df['tweet'] = tokenized_tweet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "#vocab = cv.vocabulary_\n",
    "y = train_df.iloc[:, 1].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, ..., 0, 0, 0], dtype=int64)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "7915    live loud #lol #liveoutloud #selfi #smile #son...\n",
       "7916    would like wish amaz make everi minut count #t...\n",
       "7917    help love year neighbor with ipad thi morn jus...\n",
       "7918    final #smart #pocket #wifi stay connect anytim...\n",
       "7919    appl barcelona #appl #store #bcn #barcelona #t...\n",
       "Name: tweet, dtype: object"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['tweet'].tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "bow_vectorizer = CountVectorizer(max_df=0.90, min_df=2, max_features=1000, stop_words='english')\n",
    "# bag-of-words feature matrix\n",
    "bow = bow_vectorizer.fit_transform(combi['tidy_tweet'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_bow = bow[:7920,:]\n",
    "test_bow = bow[7920:,:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(train_bow, \n",
    "                                                    y, \n",
    "                                                    test_size = 0.25, \n",
    "                                                    random_state = 101)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\ZALAKKUMARBHALANI\\Anaconda3\\envs\\deeplearning\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
       "                   multi_class='warn', n_jobs=None, penalty='l2',\n",
       "                   random_state=None, solver='warn', tol=0.0001, verbose=0,\n",
       "                   warm_start=False)"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "lr_cv = LogisticRegression()\n",
    "lr_cv.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = lr_cv.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_lr = lr_cv.predict_proba(X_test)\n",
    "y_pred_lr_int = y_pred_lr[:,1] >= 0.30\n",
    "y_pred_lr_int = y_pred_lr_int.astype(np.int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      0.88      0.92      1450\n",
      "           1       0.73      0.88      0.80       530\n",
      "\n",
      "    accuracy                           0.88      1980\n",
      "   macro avg       0.84      0.88      0.86      1980\n",
      "weighted avg       0.89      0.88      0.88      1980\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix,accuracy_score,classification_report,roc_auc_score,roc_curve\n",
    "cm = confusion_matrix(y_test, y_pred_lr_int)\n",
    "ac = accuracy_score(y_test, y_pred_lr_int)\n",
    "roc = roc_auc_score(y_test, y_pred_lr_int)\n",
    "print(classification_report(y_test, y_pred_lr_int))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_pred = lr_cv.predict_proba(test_bow)\n",
    "test_pred_int = test_pred[:,1] >= 0.3\n",
    "test_pred_int = test_pred_int.astype(np.int)\n",
    "test_df['label'] = test_pred_int\n",
    "submission = test_df[['id','label']]\n",
    "submission.to_csv('sub_lr_bow.csv', index=False) # writing data to a CSV file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,\n",
       "    decision_function_shape='ovr', degree=3, gamma=0.1, kernel='linear',\n",
       "    max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
       "    tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.svm import SVC\n",
    "svc_cv = SVC(kernel='linear', gamma=0.1)\n",
    "svc_cv.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_svc = svc_cv.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.92      0.91      0.91      1450\n",
      "           1       0.76      0.78      0.77       530\n",
      "\n",
      "    accuracy                           0.87      1980\n",
      "   macro avg       0.84      0.85      0.84      1980\n",
      "weighted avg       0.88      0.87      0.87      1980\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cm = confusion_matrix(y_test, y_pred_svc)\n",
    "ac = accuracy_score(y_test, y_pred_svc)\n",
    "roc = roc_auc_score(y_test, y_pred_svc)\n",
    "print(classification_report(y_test, y_pred_svc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_pred = model.predict(test_bow)\n",
    "# test_pred_int = test_pred[:,1] >= 0.3\n",
    "test_pred_int = test_pred.astype(np.int)\n",
    "test_df['label'] = test_pred_int\n",
    "submission = test_df[['id','label']]\n",
    "submission.to_csv('sub_ann_bow.csv', index=False) # writing data to a CSV file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import xgboost as xgb\n",
    "xgbc = xgb.XGBClassifier(learning_rate=0.05, n_estimators=500, max_depth=5, gamma=0.5)\n",
    "xgbc.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_xgb = xgbc.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cm = confusion_matrix(y_test, y_pred_xgb)\n",
    "ac = accuracy_score(y_test, y_pred_xgb)\n",
    "roc = roc_auc_score(y_test, y_pred_xgb)\n",
    "print(classification_report(y_test, y_pred_xgb))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "                       max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
       "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                       min_samples_leaf=1, min_samples_split=2,\n",
       "                       min_weight_fraction_leaf=0.0, n_estimators=1000,\n",
       "                       n_jobs=None, oob_score=False, random_state=None,\n",
       "                       verbose=0, warm_start=True)"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "rfc_cv = RandomForestClassifier(n_estimators=1000, warm_start=True)\n",
    "rfc_cv.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_rfc = rfc_cv.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.90      0.92      0.91      1450\n",
      "           1       0.76      0.72      0.74       530\n",
      "\n",
      "    accuracy                           0.86      1980\n",
      "   macro avg       0.83      0.82      0.82      1980\n",
      "weighted avg       0.86      0.86      0.86      1980\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cm = confusion_matrix(y_test, y_pred_rfc)\n",
    "ac = accuracy_score(y_test, y_pred_rfc)\n",
    "roc = roc_auc_score(y_test, y_pred_rfc)\n",
    "print(classification_report(y_test, y_pred_rfc))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_rfc = rfc_cv.predict_proba(X_test)\n",
    "y_pred_rfc_int = y_pred_rfc[:,1] >= 0.3\n",
    "y_pred_rfc_int = y_pred_rfc_int.astype(np.int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      0.86      0.90      1450\n",
      "           1       0.70      0.88      0.78       530\n",
      "\n",
      "    accuracy                           0.87      1980\n",
      "   macro avg       0.83      0.87      0.84      1980\n",
      "weighted avg       0.88      0.87      0.87      1980\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cm = confusion_matrix(y_test, y_pred_rfc_int)\n",
    "ac = accuracy_score(y_test, y_pred_rfc_int)\n",
    "roc = roc_auc_score(y_test, y_pred_rfc_int)\n",
    "print(classification_report(y_test, y_pred_rfc_int))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "dt_cv = DecisionTreeClassifier(max_depth=25)\n",
    "dt_cv.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_dt = dt_cv.predict_proba(X_test)\n",
    "y_pred_dt_int = y_pred_dt[:,1] >= 0.3\n",
    "y_pred_dt_int = y_pred_dt_int.astype(np.int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cm = confusion_matrix(y_test, y_pred_dt_int)\n",
    "ac = accuracy_score(y_test, y_pred_dt_int)\n",
    "roc = roc_auc_score(y_test, y_pred_dt_int)\n",
    "print(classification_report(y_test, y_pred_dt_int))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "tfidf_vectorizer = TfidfVectorizer(max_df=0.90, min_df=2, max_features=1000, stop_words='english')\n",
    "# TF-IDF feature matrix\n",
    "tfidf = tfidf_vectorizer.fit_transform(combi['tidy_tweet'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\ZALAKKUMARBHALANI\\Anaconda3\\envs\\deeplearning\\lib\\site-packages\\sklearn\\linear_model\\logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "train_tfidf = tfidf[:7920,:]\n",
    "test_tfidf = tfidf[7920:,:]\n",
    "\n",
    "xtrain_tfidf, xvalid_tfidf, ytrain, yvalid = train_test_split(train_tfidf, y, random_state=101, test_size=0.25)\n",
    "\n",
    "lr_cv.fit(xtrain_tfidf, y_train)\n",
    "\n",
    "prediction = lr_cv.predict_proba(xvalid_tfidf)\n",
    "prediction_int = prediction[:,1] >= 0.3\n",
    "prediction_int = prediction_int.astype(np.int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      0.87      0.91      1450\n",
      "           1       0.71      0.88      0.79       530\n",
      "\n",
      "    accuracy                           0.87      1980\n",
      "   macro avg       0.83      0.88      0.85      1980\n",
      "weighted avg       0.89      0.87      0.88      1980\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cm = confusion_matrix(yvalid, prediction_int)\n",
    "ac = accuracy_score(yvalid, prediction_int)\n",
    "roc = roc_auc_score(yvalid, prediction_int)\n",
    "print(classification_report(yvalid, prediction_int))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}