A comparative study of the class imbalance problem in Twitter spam detection

摘要

Recently, online social network (OSN) such as Twitter has become an important and popularrnsource for real-time information and news dissemination, and Twitter is inevitably a prime targetrnof spammers. It has been showed that the security threats caused by Twitter spam can reach farrnbeyond the social media platform itself. To mitigate the damage caused by Twitter spam,machinernlearning classification algorithms have been employed by researchers and communities to detectrnthe Twitter spam.However,most of these studies have overlooked the class imbalance problem inrnTwitter spamdetection. In this paper,wehave studied the class imbalance problem inTwitter spamrndetection. Firstly, we have conducted a comparative study regarding some popular methods inrnhandling the class imbalance problem in order to identify themost effective approach for addressingrnthe class imbalance problem. Then, we have conducted another comparative study fromrnTwitter spam detection based on several classic techniques. Experimental results demonstraternthat a fuzy-based ensemble learning can significantly improve the classification performance onrnimbalance ground truth Twitter data.