The Implementation of Naïve Bayes Algorithm for Classifying Tweets Containing Hate Speech with Political Motive

The mid-of 2018 until the mid-of 2019 has been densely marked with political agendas in Indonesia. This moment has created vulnerability to spread hate speech with political motive which is one of the most commonly encountered as cyber-crime on Indonesia's social media. Twitter, as one of the most popular social medias in Indonesia becomes the target of spreading hate speech. Tweets that are suspected to contain hate speech can be drawn automatically using twitter scraper. Filtering and labeling stage before being classified using Naïve Bayes Algorithm. The classification process is done by using WEKA, so finally the accuracy of Naïve Bayes Algorithm for tweet containing hate speech with political motive can be identified. By using that method, the classification can be started and the particular tweets that include non-political hate speech, hate speech with political motive, or non-hate speech can be identified. Data tweet that has been drawn, should then be processed through the average value from the accuracy is 93.4%.