Sentiment Analysis of Tweets: A Machine Learning Approach

The growth and advancement in social network platforms increase the number of users noticeably. Social network platforms, like Twitter, grant users the ability to interact and express their emotions about events. Since Twitter platform involves all ages with a fair representation of gender, the sentiment analysis of Twitter data reflects the general feelings of people about a particular event. The sentiment analysis is a natural language processing (NLP) method that mainly focuses on deciding whether the sentiment is positive, negative, or neutral. Additionally, it is referred to as material polarity or mining of opinions. In the context of sentiment analysis, various approaches can be applied such as the Lexicon and machine learning (ML) approaches. Compared with lexicon approach, ML approach is considered simple and more efficient. In this study aims at Performing sentiment analysis of Twitter data related to COVID19 using the ML approach. Four ML models are used in this study namely, linear support vector classification (Linear SVC), logistic regression (LR), decision tree (DT), and random forest (RF). The performance of the above-mentioned models is tested using various metrics such as accuracy, recall, precision, and F1 score. The results released that the Linear SVC model has superior performance among the other models.