The main goal of the project is to be able to accurately categorize a tweet into positive, negative or neutral, based on training the data that is already categorized. Initially, I performed data cleaning and preprocessing and then exploratory data analysis to understand the data better. Finally, I applied various machine learning models, of which Naive Bayes performed best, with 67.50% accuracy.
Dataset: Cleaning tweets and exploratory data analysis on "Sentiment140 dataset with 1.6 million tweets" dataset from kaggle (Dataset link: https://www.kaggle.com/kazanova/sentiment140)