Today, Tweet is used often and has significant meaning in the lives of many people, including businesspeople, media professionals, politicians, and others. Twitter, one of the most widely used social networking services, allows its users to express their views on a wide variety of topics, including politics, sports, the economy, pop culture, and more. It's one of the quickest ways to share data with others. The way people think is profoundly impacted by it. Twitter has become a breeding ground for criminals who use anonymity to commit crimes. Recognizing Twitter bots is crucial because of the potential threat they represent to other users. Thus, it is essential that tweets be posted by actual individuals and not by Twitter bots. The Twitter feed is being spammed by a bot. Therefore, recognizing bots helps in recognizing spam communications. Features extracted from Twitter accounts are utilized by machine learning algorithms to determine whether or not a user is authentic. In this research, we used three different machine learning techniques—a Decision Tree, a Random Forest, and a Multinomial Naive Bayes—to establish whether or not an account was genuine. We evaluate the classification performance and accuracy of several methods. Approximately 89% accuracy is achieved by the Multinomial Naive Bayes approach, 90% by the Random Forest algorithm, and 93% by the Decision Tree method. It follows that Decision tree achieves higher accuracy than both Random Forest and Multinomial Nave Bayes.