A New Classification Method to Overcome Over—Branching

Classification is an important technique in data mining.The decision trees builty by most of the existing classification algorithms commonly feature over-branching,which will lead to poor efficiency in the subsequent classification period.In this paper,we present a new value-oriented classification method,which aims at building accurately proper-sized decision trees while reducing over-branching as much as possible,based on the concepts of frequent-pattern-node and exceptive-child-node.The experiments show that while using relevant anal-ysis as pre-processing ,our classification method,without loss of accuracy,can eliminate the over-branching greatly in decision trees more effectively and efficiently than other algorithms do.

A new classification method to overcome over-branching

Classification is an important technique in data mining. The decision trees built by most of the existing classification algorithms commonly feature over-branching, which will lead to poor efficiency in the subsequent classification period. In this paper, we present a new value-oriented classification method, which aims at building accurately proper-sized decision trees while reducing over-branching as much as possible, based on the concepts of frequent-pattern-node and exceptive-child-node. The experiments show that while using relevant analysis as pre-processing, our classification method, without loss of accuracy, can eliminate the over-branching greatly in decision trees more effectively and efficiently than other algorithms do. This work is partially supported by the NKBRSF of China (G1998030414) and the National Natural Science Foundation of China (No.60003016). ZHOU Aoying received his M.S. degree in computer science from Sichuan. University in 1988, and his Ph.D. degree in computer software from Fudan University in 1993. He is currently a professor in the Department of Computer Science, Fudan University. His main research interests include object-oriented data model for multimedia information, Web data management, data mining and data warehousing, the novel database technologies and their applications to digital library and electronic commerce. QIAN Weining is a Ph.D. candidate in the Departemnt of Computer Science, Fudan University. His speciality is database and knowledge base. He is supported by Microsoft Research Fellowship. His research interests include clustering, data mining and Web mining. QIAN Hailei is a graduate student in the Department of Computer Science Fudan University. Her speciality is database and knowledge base. Her research interests include clustering and data mining. JIN Wen is a Ph.D. candidate in the School of Computing, Simon Fraser University, Canada, supervised by Dr. Jiawei Han. His current research interests are database and data warehousing, data mining, Web mining and XML.