Discovering High-quality Users from Sina Weibo Based on Trust Transfer Model ?

This paper devotes to discovering the high-quality users from Sina microblog (Weibo) which is the most popular microblog site in China. First, the Trust Transfer Model (TTM) is introduced as a theoretical background to make sure that users are trustworthy and high-quality. Then, a Breadth First Search (BFS) crawler based on TTM is implemented to capture users’ profile data via Weibo APIs. Thereupon, with the help of the BFS crawler, a dataset filled with more than 20 million high-quality users is constructed. Furthermore, analysis and discussion are presented in detail to prove the users in the dataset are high-quality, and Coincidence Ratio (CR) is proposed as a notable indicator for watching the trend of high-quality users captured by the crawler with the time goes by. Finally, an evaluation based on CR indicates the high accordance between our data and official data; meanwhile, testing for the degree of coverage based on the “Top10 Hot Microblogs” in Weibo presents a relatively high credibility of our dataset. In conclusion, we will continue efforts to discover the high-quality users in microblog and believe that discovering and maintaining a dataset filled with the high-quality users which is not too big is quite significant for both academic research and business application.