报告题目：Event Detection and Multi-Dimensional Analysis based on Multi-Modal Big Data
报 告 人：李 青教授（香港城市大学）
Qing Li is a Professor at the Department of Computer Science, and the Director of the Engineering Research Centre on Multimedia Software at the City University of Hong Kong, where he joined as a faculty member since Sept 1998. He received his B.Eng. from Hunan University (Changsha), and M.Sc. and Ph.D. degrees from the University of Southern California (Los Angeles), all in computer science. His research interests include multimedia retrieval and management, conceptual data modeling, social media and Web services, and e-learning systems. He has authored/co-authored over 300 publications in these areas. He is actively involved in the research community and has served as an associate editor of a number of major technical journals including IEEE Transactions on Knowledge and Data Engineering (TKDE), ACM Transactions on Internet Technology (TOIT), Data & Knowledge Engineering (DKE), World Wide Web (WWW), and Journal of Web Engineering, in addition to being a Conference and Program Chair/Co-Chair of numerous major international conferences. He also sits in the Steering Committees of ACM RecSys, DASFAA, ER, ICWL and IEEE U-MEDIA. Prof. Li is a Fellow of IET (UK), a senior member of IEEE (US) and a distinguished member of CCF (China).
The publicly available data such as the massive and dynamically updated news and social media data streams (a.k.a. big data) covers the various aspects of social activities, personal views and expressions, which points to the importance of understanding and discovering the knowledge patterns underlying the big data, and the need of developing methodologies to this end, particularly to discover real-world events from such big data. In this talk we aim to provide insight on event detection and prediction underlying the Big Data. We propose to develop techniques for untargeted event detection (UED) and targeted event detection (TED) from multi-sourced data, and build event cube models to support various event queries and analysis thereon. UED aims to discover numerous varieties of real-world events while TED aims to discover domain-specific events. By discovering various kinds of real-world events from the big data, UED supports decision-makings by providing the newest and comprehensive knowledge about the events. Under many circumstances, the events discovered by UED may actually serve as the “context” of those discovered by TED which aims to foresee some domain-specific events and allows users to make responses in the first place. Specifically, active and passive data collecting will be employed to obtain the streaming data from multiple data domains, which will be cleaned by removing noisy and inconsistent data through crowd-sourcing techniques. To learn accessible data representations, multimodal fusion models are devised based on dictionaries to be constructed over the event elements. Furthermore, clustering-based and classification-based event discovery models are designed, respectively, for UED and TED tasks. To organize the discovered events and to facilitate mining the inherent relationships among them, we also devise an event cube model according to the multiple dimensions of event elements.