Sentiment Analysis_Final Solution_0.5230949840.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-info\">\n",
    "This is <b>Rank 29</b> solution using SGD Classifier and One Vs Rest Classifier. \n",
    "  \n",
    "I have also tried <b>fastai</b> but it didn't give me good results even after fine-tuning it. I have also uploaded fastai solution in this folder which gave me F1-Score of 0.4676695183 on Public leaderboard which will be good reference point for using fastai for NLP problems.\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Import Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "import nltk\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import warnings \n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nltk.corpus import stopwords\n",
    "nltk_stopwords = stopwords.words('english')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Import Datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "train = pd.read_csv('../input/innoplexus/innoplexus_sentiment analysis/train.csv')\n",
    "test = pd.read_csv('../input/innoplexus/innoplexus_sentiment analysis/test.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No. of samples in training set: 5279\n",
      "No. of samples in test set    : 2924\n"
     ]
    }
   ],
   "source": [
    "print('No. of samples in training set:', len(train))\n",
    "print('No. of samples in test set    :', len(test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Text pre-processing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def clean_text(text):\n",
    "    text = text.lower()\n",
    "    text = re.sub(r'@[a-zA-Z0-9_]+', '', text)   \n",
    "    text = re.sub(r'https?://[A-Za-z0-9./]+', '', text)   \n",
    "    text = re.sub(r'www.[^ ]+', '', text)  \n",
    "    text = re.sub(r'[a-zA-Z0-9]*www[a-zA-Z0-9]*com[a-zA-Z0-9]*', '', text)  \n",
    "    text = re.sub(r'[^a-zA-Z]', ' ', text)   \n",
    "    text = [token for token in text.split() if len(token) > 2]\n",
    "    text = ' '.join(text)\n",
    "    return text\n",
    "\n",
    "train['text'] = train['text'].apply(clean_text)\n",
    "test['text'] = test['text'].apply(clean_text)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Train test split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_cv, y_train, y_cv = train_test_split(train['text'], train['sentiment'], test_size=0.25, stratify=train['sentiment'], \n",
    "                                                random_state=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature extraction & Model building"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "tfidf_vect = TfidfVectorizer(analyzer='word', stop_words=None, token_pattern='(?u)\\\\b\\\\w\\\\w+\\\\b', ngram_range=(1, 1), max_df=1.0, \n",
    "                             min_df=3, max_features=None, binary=False, norm='l2', use_idf=True, smooth_idf=True, sublinear_tf=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_tfidf = tfidf_vect.fit_transform(X_train)\n",
    "X_cv_tfidf = tfidf_vect.transform(X_cv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "f1_score       : 0.5145775153047341\n",
      "accuracy score : 0.6613636363636364\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import f1_score, accuracy_score\n",
    "from sklearn.multiclass import OneVsRestClassifier\n",
    "from sklearn.linear_model import SGDClassifier\n",
    "sgd = SGDClassifier(loss='log', max_iter=200, random_state=0, class_weight='balanced')\n",
    "ovr = OneVsRestClassifier(sgd)\n",
    "ovr.fit(X_train_tfidf, y_train)\n",
    "y_pred_class = ovr.predict(X_cv_tfidf)\n",
    "print('f1_score       :', f1_score(y_cv, y_pred_class, average='macro'))\n",
    "print('accuracy score :', accuracy_score(y_cv, y_pred_class))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Predict on test set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "tfidf_vect = TfidfVectorizer(analyzer='word', stop_words=None, token_pattern='(?u)\\\\b\\\\w\\\\w+\\\\b', ngram_range=(1, 1), max_df=1.0, \n",
    "                             min_df=3, max_features=None, binary=False, norm='l2', use_idf=True, smooth_idf=True, sublinear_tf=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "full_text = list(train['text'].values) + list(test['text'].values)\n",
    "tfidf_vect.fit(full_text)\n",
    "\n",
    "X_train_tfidf = tfidf_vect.transform(train['text'])\n",
    "X_test_tfidf = tfidf_vect.transform(test['text'])\n",
    "\n",
    "y_train = train['sentiment']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([2, 1, 2, ..., 2, 2, 2])"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.multiclass import OneVsRestClassifier\n",
    "from sklearn.linear_model import SGDClassifier\n",
    "sgd = SGDClassifier(loss='log', max_iter=200, random_state=0, class_weight='balanced')\n",
    "ovr = OneVsRestClassifier(sgd)\n",
    "ovr.fit(X_train_tfidf, y_train)\n",
    "y_pred_class = ovr.predict(X_test_tfidf)\n",
    "y_pred_class"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Submission"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>unique_hash</th>\n",
       "      <th>sentiment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9e9a8166b84114aca147bf409f6f956635034c08</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>e747e6822c867571afe7b907b51f0f2ca67b0e1a</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>50b6d851bcff4f35afe354937949e9948975adf7</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7f82ec2176ae6ab0b5d20b5ffc767ac829f384ae</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>8b37d169dee5bdae27060949242fb54feb6a7f7f</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                unique_hash  sentiment\n",
       "0  9e9a8166b84114aca147bf409f6f956635034c08          2\n",
       "1  e747e6822c867571afe7b907b51f0f2ca67b0e1a          1\n",
       "2  50b6d851bcff4f35afe354937949e9948975adf7          2\n",
       "3  7f82ec2176ae6ab0b5d20b5ffc767ac829f384ae          2\n",
       "4  8b37d169dee5bdae27060949242fb54feb6a7f7f          2"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test['sentiment'] = y_pred_class\n",
    "test.drop(['text','drug'], axis=1,inplace=True)\n",
    "test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2    1958\n",
       "0     487\n",
       "1     479\n",
       "Name: sentiment, dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test['sentiment'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "test.to_csv('submission.csv', index=False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}