Twitter Sentiment Analysis using Machine Learning and Deep Learning Techniques
Vedant Pandya, Anuja Somthankar, Shivanshu S Shrivastava, Megharani Patil
Abstract
Nowadays, Social Media websites have become increasingly popular, hosting millions of users daily and accounting for countless interactions. Among them, Twitter is extraordinarily popular, boasting a huge user base and a preeminent global presence. People from all walks of life use Twitter to express their opinions on a myriad of topics. Thus, by employing Sentiment Analysis techniques on tweets, Organizations can get valuable insights on public opinions, which can allow them to make more effective decisions. In this paper, Machine Learning and Deep Learning models were trained on a dataset containing 1.6 million tweets, to classify them into positive or negative sentiments. The performance of these models was compared with respect to metrics like Accuracy, F1-Score, Recall and Precision.