Implementation of K-Nearest Neighbor For Classify Hate Speech on Twitter

Twitter is a social networking platform that uses microblogging. A form of social media called microblogging enables individuals to write and publish actions or ideas without restriction. Twitter's effects are mixed, including both positive and negative ones. The result is the rise of various criminal offenses, such as hate speech. Of course, where it can take a long time to be discovered by the algorithm, it requires the assistance of a linguist to identify hate speech. The K-Nearest Neighbour approach is used in this study, with a total of 1000 data points split into 300 test points and 700 training points. with the distinction between racial, racist and neutral hate speech. The accuracy of the training data increased by 70%, 80%, and 90% according to the testing process using the confusion matrix, or 98.5%, 98.3%, and 98.5% with a value of k=7. The majority of testing data is categorised as 33% religious hate speech, 17% racial hate speech, and 50% neutral.