available to internet users thanks to the development and growth of online technologies. The internet has developed into a forum for online education, idea sharing, and opinion exchange. Social networking services like Twitter, Facebook, and Google+ are quickly gaining popularity as a result of the ability for users to share and express their opinions on many subjects, engage in conversation with various communities, and broadcast messages globally. The study of sentiment in Twitter data has received a lot of attention. This study primarily focuses on sentiment analysis of Twitter data, which is useful for analyzing information in tweets when opinions are very unstructured, varied, and occasionally neutral. The strategies for opinion mining that are currently in use, such as lexicon-based approaches and machine learning, are surveyed, compared, and evaluated in this work along with evaluation measures. We present research on Twitter data streams using different machine learning techniques such as Naive Bayes, Max Entropy, and Support Vector Machine. Additionally, we covered the general difficulties and uses of sentiment analysis on Twitter.