University of Minnesota
University Relations
http://www.umn.edu/urelate
612-624-6868
myU OneStop


Go to unit's home.

Home | Seminars and Symposia | Past seminars/symposia: Wednesday, October 29, 2003

DTC Seminar Series

Efficient Mining of Closed Patterns with Tough Constraint

by

Jianyong Wang
University of Minnesota
Digital Technology Center

Wednesday, October 29, 2003
1:00 pm

402 Walter Library

Jianyong WangDownload slides (pdf 1.43 MB) Frequent pattern mining is a fundamental issue in data mining research and various efficient algorithms have been proposed in the past 10 years. The problem with these algorithms is that they may generate an exponentially large number of frequent patterns when the support is low. To overcome this, mainly two classes of techniques have been proposed recently. The first focuses on mining closed/maximal patterns, while the second class tries to incorporate various anti-monotone, monotone, or convertible constraints into frequent pattern mining. A limitation with these two kinds of techniques is that they use a constant support value, irrespective of the length of the discovered patterns. In this talk we will specifically discuss how to push deeply the length-decreasing support constraint into closed itemset mining, which is a particularly challenging problem due to the fact that the downward-closure property cannot be used to prune the search space. We will first summarize the recent progress on closed pattern mining, then focus on several newly proposed pruning methods which can push deeply the length-decreasing support constraint into closed itemset mining, and introduce an efficient algorithm, BAMBOO. Finally we will also present some performance study.

Jianyong Wang presenting

 

Jianyong Wang received his Ph.D. in Computer Science in 1999 from Institute of Computing Technology, the Chinese Academy of Sciences. Since then he has worked in Peking University in the areas of distributed systems and web search engines, and visited Simon Fraser University (Canada) and University of Illinois as a postdoc research fellow, mainly working in the area of data mining. After joining the Digital Technology Center as a research associate in July 2003, he has been mainly working with Prof. George Karypis's group on some chanllenging data mining problems.