[Adapt] FW: Talk by Professor Jie Wang from University of Massachusetts, USA
Kenny Zhu
kzhu at cs.sjtu.edu.cn
Mon Mar 12 23:02:27 CST 2012
Hi guys,
Please go to this talk which seems to be quite related to our work, especially those of you who work on information extraction/knowledge discovery.
K
-----Original Message-----
From: Xiaofeng Gao [mailto:gao-xf at cs.sjtu.edu.cn]
Sent: 2012年3月12日 22:43
To: all at cs.sjtu.edu.cn
Cc: wang at cs.uml.edu
Subject: Talk by Professor Jie Wang from University of Massachusetts, USA
Dear colleagues,
Prof. Jie Wang from University of University of Massachusetts, Lowell, USA, is visiting our department on March 15th, 2012. He is the Chair and Professor of Computer Science at UML, and he will present a talk related to knowledge network. The following are details of his talk. You are welcome to attend this presentation. Please also disseminate this announcement to your students who might be interested.
Thanks,
Xiaofeng Gao
--------------------------------------------
Title: Finding Statistical Characteristics and Similarities of Substructures between Knowledge Networks
Time: 3:00PM, Thursday, March 15th, 2012
Venue: SEIEE-03-410
Speaker: Jie Wang, Chair and Professor of Computer Science, University of University of Massachusetts, Lowell, USA.
Abstract:
Statistical characteristics of knowledge networks may help reveal internal structures of knowledge and how knowledge is evolved and organized. We analyze statistical properties of domain knowledge networks and show the goodness of fit of double Pareto lognormal distribution on node degrees. Finding structural similarities between large inter-domain knowledge networks using computational methods is expected to facilitate the discovery of new knowledge. We devise an efficient algorithm using random walkers and time series to identify structural similarities of sub-networks between inter-domain knowledge networks. Concepts with similar structures identified by the algorithm may indicate unknown relationships between them, which may be worth the effort of domain experts to investigate. Our method can be applied to large knowledge repository such as Wikipedia. In particular, we devise an efficient mechanism to extract domain knowledge networks from Wikipedia, and use it to extract four domain networks, namely, mathematics, physics, biology, and chemistry. We examine these knowledge networks and list concept pairs with high similarity scores. We present strong evidence to some of these pairs that they are indeed related.
This is joint work with Weibo Gong, Zheng Fang, and Benyuan Liu.
Bio:
Dr. Jie Wang is Professor and Chair of Computer Science at the University of Massachusetts, Lowell, USA. He is also Director of China Partnerships under Provost, Director of the University Center for Network and Information Security, and co-Director of the University Center for Cyber Forensics. He received his PhD in Computer Science from Boston University in 1991, Master of Engineering in Computer Science from Zhongshan (Sun Yat-sen) University in 1985, and Bachelor of Science in Computational Mathematics from Zhongshan University in 1982. His research interests include computational complexity theory, modeling and algorithms, network security, and computational medicine. He has worked as a security consultant in financial industry. His recent research focus is on network dynamics, knowledge discovery, and wireless sensor networks. His research has been funded by the NSF since 1991. IBM, Intel, Google and the Natural Science Foundation of China have also funded his research. He has published over 150 research papers in some of the most prestigious journals and conference proceedings. He has authored and co-authored five books; edited and co-edited four books. He is active in professional service, including chairing conference program committees, serving as editor-in-chief or a book series on modeling and algorithms and as journal editors, and organizing conferences and workshops.
More information about the Adapt
mailing list