The concept of "dominance" is an important concept in skyline computation. In the NICHE project, our aim is to apply this concept of dominance for knowledge discovery. Our studies focused on two directions. First, we look at the processing of skyline in high dimensional space (around 10-50 dimensions in our studies). This is essential in data mining since high dimensional datasets are the one that give the most challenges to data analysis most of the time. Furthermore, we feel that skyline is the much more important in high dimensional space where users have much more dimensions of choice compare to small number of dimensions where user's preference are usually much clearer. Second, we look at the concept of skyline from the product owner perspective. Instead of asking where are the skylines, we look at how products can become part of the skyline while remaining profitable. This involve both the analysis of both users and competitor products and also customer preference. In turn, we also look at game theory where products are pit against each other to "dominate" as many customers as possible. Our studies show that Nash equilibrium can be achieved in such a game and our algorithm cover at least half the customers with some products compared to an optimal algorithm.
Anthony K. H. Tung
Zhengjie Zhang
Beng Chin Ooi
Kian-Lee Tan
Chee-Yong Tan
Martin Ester
Laks V. S. Lakshmanan
Cuiping Li
H. V. Jagadish
Wen Jin
Zhenjie Zhang, Laks V. S. Lakshmanan, Anthony K. H. Tung. "On Domination Game Analysis for Microeconomic Data Mining". To appear in Transactions on Knowledge Discovery from Data (TKDD), Volume 2, Issue 4, 2008.
Cuiping Li, Anthony K. H. Tung, Wen Jin, Martin Ester. "On Dominating Your Neighborhood Profitably". In the 33rd Very Large Data Bases conference (VLDB'07), Vienna, 2007
Cuiping Li, Beng Chin Ooi, Anthony K. H. Tung, Shan Wang. "DADA: A Data Cube for Dominant Relationship Analysis", In ACM SIGMOD Int'l. Conference on Management of Data , Chicago, 2006 (SIGMOD'06).
Chee-Yong Chan, H. V. Jagadish, Kian-Lee Tan, Anthony K. H. Tung, Zhenjie Zhang. "Finding k-Dominant Skylines in High Dimensional Space", In ACM SIGMOD Int'l. Conference on Management of Data , Chicago, 2006 (SIGMOD'06).
Wen Jin, Anthony K. H. Tung, Martin Ester and Jiawei Han,''On Efficient Processing of Subspace Skyline Queries on High Dimensional Data'', in Proc. 2007 Int'l.Conf.on Scientific and Statistical Database Management (SSDBM'07), Canada, July. 2007.
Zhenjie Zhang, Xinyu Guo, Hua Lu, Anthony K.H. Tung, and Nan Wang. "Discovering Strong Skyline Points in High Dimensional Spaces".(Poster) in CIKM 2005.