Twitter Sentiment Analysis

# Library to work with regular expressions, required during cleaning the data

import re

# Working with numpy arrays during data exploration, like count

import numpy as np

# Used to store the csv data in data frames and perform various operations like groupby etc

import pandas as pd

# The below given libraries are used for plotting various graphs during data exploration

import seaborn as sns

from wordcloud import WordCloud

import matplotlib.pyplot as plt

# The below given library is used for stemminga dn lemmatization

from nltk.stem import WordNetLemmatizer

# Scikit Learn is the library used to apply various algorithms as shown below

# Import used for Support Vector Machine Library

from sklearn.svm import LinearSVC

# Import to implement Beroulli Naive Bayes Algorithm for classification

from sklearn.naive_bayes import BernoulliNB

# Logistic regresssion for classification

from sklearn.linear_model import LogisticRegression

# Import to split the complete dataset into train and test data

from sklearn.model_selection import train_test_split

# Import used for working on Term Frequency and Inverse Docuemnt Frequency

from sklearn.feature_extraction.text import TfidfVectorizer

#Code used to asses the algorithm using Accuracy, F1Score etc, and also used to build the confusion matrix. To draw the AUC-ROC Curve

from sklearn.metrics import confusion_matrix, classification_report

# Stemming

import nltk

st = nltk.PorterStemmer()

def stemming_on_text(data):

text = [st.stem(word) for word in data]

return text

dataset['text']= dataset['text'].apply(lambda x: stemming_on_text(x))

dataset['text'].head()

# Lemmatization

import nltk

nltk.download('wordnet')

lm = nltk.WordNetLemmatizer()

def lemmatizer_on_text(data):

text = [lm.lemmatize(word) for word in data]

final_text=" ".join([word for word in text])

return final_text

dataset['text'] = dataset['text'].apply(lambda x: lemmatizer_on_text(x))

dataset['text'].head()