Associate Professor Guo Xunhua from Tsinghua University paid an exchange visit to our school. At December 18, 2017, at 10:00 am in the conference room of main building, Prof. Guo gave an academic report entitled "Calibrating the Helpfulness of Online Product Reviews: An Iterative Bayesian Probability Approach ". The report was hosted by Professor Yan Zhijun. A number of  teachers and students attended the meeting.

　　Voting mechanisms are widely used to evaluate the quality of user-generated content, such as online product reviews. However, the existing methods mainly focus on the analysis of functional characteristics, ignoring the frequency of user voting and the information conveyed by the mode. Based on the prediction theory and consumer behavior research, Assoc Prof Guo Xunhua proposed an iterative Bayesian distribution estimation method to predict more accurately the usefulness of the model training set. The study used Amazon's online commentary data and used a simulation test and a two-phase data test to verify the accuracy of the method. In addition, the study conducted an out-of-sample user study that again demonstrated the predictive power of the new method. This approach helps users tap more useful information and offers new insights into referral systems and social media analytics.