Measuring credibility of users in an E-learning social network

Learning Villages (LV) is an E-learning platform for people´s online discussions and frequently citing postings of one another. It will greatly improve learning efficiency if credible users can be accurately identified in the E-learnning community. In this paper, we propose a novel method to rank credibility of users in the LV system. We first propose a k-EACM graph to describe the article citation structure in the LV system. The k-EACM graph not only describes the citation attitude between articles but also takes into account indirect citation links. We further build a weighted graph model k-UCM graph to reveal the implicit relationships between users hidden behind the citations among their articles. Finally, we design a graph based ranking algorithm, called Credible Author Ranking (CAR) algorithm, which can be applied to rank nodes in a graph with negative edges. We perform experiments on three simulated data sets. The experimental results show that our proposed method works well to rank credibility of users in the LV system. The comparison of results on average credibility of users in each set and composition of users in top N of three cases demonstrates that the CAR algorithm ranks users credibility in consistent with the predefined ground truth.