张仲非

报告题目：Uni-party and bi-party community generation

报告人：张仲非教授

报告时间：2005年11月22日

摘要：

Community generation plays an important role in mining diverse data collections in many applications. In this talk, I will give an overview on the theory of community generation that is being developed in the Multimedia Research Lab in SUNY Binghamton. Specifically, I will discuss the issues related to uni-party data and bi-party data community generation. For uni-party data community generation, I will discuss the general methodolody of link discovery based on correlation analysis as well as its identified applications.For bi-party data community generation, I will discuss the block value decomposition method and the soft ensemble techniques. I will also discuss the potential applications of this research.

个人简介：

张仲非教授是国际计算机视觉、模式识别、数据挖掘界的著名学者, 目前是State University of New York at Binghampton的Associate Professor, 多媒体研究室主任.他早年在浙江大学获学士、硕士学位，在美国University of Massachusetts at Amherst 计算机科学系获博士学位，博士毕业后曾在State University of New York at Buffalo 任Assistant Professor，曾在美国国家科学院、空军研究实验室、微软研究院等机构从事访问研究，目前是著名国际刊物的编委、多次担任国际会议程序委员，并担任美国政府和私人一些科研基金会的评审专家组成员。