|| With the rapid development of data acquisition, IT infrastructures,social networks and Web2.0 applications, more and more massive, heterogeneous, uncertain,dynamically changing and socialized data are generated and stored in distributedsystems. Discovering implied knowledge from data is always the topic with greatattention for data understanding, data utilization and information services,where uncertainty is ubiquitous. We adopt Bayesian network (BN), one of thepopular and important probabilistic graphical models, as the effectiveframework for representing and inferring uncertain knowledge by means ofqualitative and quantitative manners. In this report, we present our studies onacquisition, representation, inference and fusion of uncertain knowledgeimplied in data, oriented to the applications of provenance analysis and userpreference modeling.
|uาะ๎(Introduction of Lecturer)F
|| Dr. Kun Yue is a professor of computer science, the vice dean of Schoolof Information Science and Engineering at Yunnan University. He received hisM.S. degree in computer science from Fudan University in 2004, and received hisPh.D. degree in computer science from Yunnan University in 2009. His researchinterests include massive data analysis, artificial intelligence, and knowledgeengineering. He is the director of Chinese Association for Artificial Intelligence(CAAI), vice Chairman designate of CAAI Uncertainty in Artificial IntelligenceSociety, and a member of Database Society of China Computer Federation (CCF).