Twitter Sentiment Analysis and Text Analytics

This project involves analyzing sentiments in tweets. The project begins with a Twitter dataset obtained from Kaggle and proceeds to clean the data using the tweet-preprocessor library and regular expressions. A 70/30 train-test split is performed, followed by vectorizing the tweets using CountVectorizer. The project then builds a Support Vector Classifier model, achieving an impressive 95% accuracy. The project aims to provide insights into sentiment analysis on Twitter data, enabling the classification of tweets as either associated with racist or sexist sentiments (labeled '1') or not'