Big Data approache to determine tweet’s polarities and find out the matching patterns, impactful tweets and to identify existing trends and predict the upcoming trends.
Data set : https://www.kaggle.com/barishasdemir/tweets-about-distance-learning
Steps Followed:
Data cleansing and preprocessing
Tokenization and Lemmatization
Use pycountry library
Use Tf-idf matrix
Use Sentiment function of textblob
Label Encoding
Create Word Cloud
Build machine learning model using Naive Bayes Classifier