Sentiment Polarity Analysis of Twitter Data Using Machine Learning Models

We live in the age of the internet which has brought drastic changes in the lives of people, along with how they perceive the world, express their opinions, and generate and manage their data over the internet. The internet is now popularly known for providing a platform for ideas to get exchanged, voices to be heard, and skills to be worked upon. Social networking sites are the most rapidly growing industry of this platform providing category. With the increase in generation of data, views, and opinions, comes up the need of sentiment analysis. In this paper, we have worked upon an available dataset of Twitter where tweets were present in a highly unstructured way including hashtags, emoticons, etc. We have researched upon the fluctuations of accuracy in predicting the polarity of tweets using different machine learning algorithms such as support vector machine (SVM) and random forest. The future hurdles for sentiment analysis on Twitter have also been discussed.