在中国科技大学获得博士学位，师从陈国良院士。目前，他的团队在数据挖掘、社会计算和普适计算等领域展开创新性的研究。他在国际会议和学术期刊上发表了200余篇学术论文，共被引用15000余次，1999年获首届微软学者奖，多次在KDD、ICDM等顶级会议上获最佳论文奖，并被邀请在ASONAM 2017、Mobiquitous 2016、SocInfo 2015、W2GIS 2011等会议做大会主题报告。他是ACM、IEEE高级会员和计算机学会杰出会员，多次担任顶级国际会议程序委员会委员和领域主席等职位。
Due to the explosive growth of information, recommender systems have been playing an increasingly important role in online services. However, these systems suffer from limited performance when data are very sparse, which is common in many application scenarios. To address this issue, auxiliary information is usually used to boost the performance. In this talk, we will introduce our recent research works which try to integrate different types of auxiliary information into recommendation systems. First, we investigate how to leverage the heterogeneous information in a knowledge base. We design three components to extract items’ semantic representations from structural content, textual content and visual content, respectively. And we propose an integrated framework, which is termed as Collaborative Knowledge Base Embedding, to jointly learn the latent representations in collaborative filtering as well as items’ semantic representations from the knowledge base. Second, we observe that people often use multiple platforms to fulfill their different information needs. Therefore, we propose a semi-supervised transfer learning method for cross-platform behavior prediction, called XPTRANS. It fully exploits the small number of overlapped crowds to optimally bridge a user’s behaviors in different platforms.