An Improved Model for Spam User Identification

Abstract

Aiming at the problem of spam user identification in microblog, this paper proposes an improved model based on representation learning and C4.5 decision tree. The spam user identification rule has been decided at first. Then, the ten-fold cross-validation and training set are combined to train the model. The experimental results show that the improved model proposed in this paper is superior to the traditional model not only in accuracy but also universality. Results of this paper may provide a guidance to the improvement of framework of a certain social network platform.