多特征微博垃圾互粉检测方法研究

Spam mutual concerns in micro-blogs bring considerable challenges to the determination of influence of bloggers and user relation strength. A method to detect the spam mutual concerns is presented. Based on features extracted from user information, relation among users, user activity, a system is trained to distinguish spammers from normal users. Experiments on real data from Sina Weibo (Micro-blog) show that our proposed method is effective with a distinguishing rate of above 80%.