Detection of False Weibo Repost Based on XGBoost
Yixin Liu, Xinhua Wang, Wen Long · 2019
This paper focuses on the false repost problem in Weibo, and the XGBoost classification algorithm is introduced to detect the true or false weibo repost and compared with some classic approaches and neural network. The model results show that XGBoost algorithm has higher classification performance than support vector machine, random forest, logistic regression and MLP. The two most important classification features of false repost are the number of weibos and the number of followers.