TWITTER BUZZER DETECTION SYSTEM USING TWEET SIMILARITY FEATURE AND SUPPORT VECTOR MACHINE

Over the past few years, people have been able to get and share information through social media easily. Some of that can be a false issue created by buzzer account intends influence into specific opinion. Politicians often use maintain good image in society utilizing accounts. The main characteristic is they upload same content repeatedly within certain period. Before analyzing data taken from such as Twitter, we need detection system filter users. This research attempts build using text processing classification method. We similarity tweets feature for applying Cosine Similarity Term Frequency - Inverse Document (TF-IDF) tweets. In addition, will other features number followers, followings, intensity tweets, ratio retweets, contain links additional this study. uses these inputs Support Vector Machine model determine whether an or not. has promising results having 89% accuracy, 86.67% precision, 70.91 % recall, 78% F1-score.