2基于奇异值分解的异常切片挖掘方法.doc

 PAGE 1288 Journal of Software 软件学报 2005,16(7)??? 遇辉 等:基于奇异值分解的异常切片挖掘  PAGE 1287 Vol.16, No.7 ?2005 Journal of Software 软 件 学 报 1000-9825/2005/16(07)1282 基于奇异值分解的异常切片挖掘? Supported by the National Natural Science Foundation of China under Grant Nos(国家自然科学基金) 作者简介: 遇辉(1977?),女,黑龙江哈尔滨人,博士生,主要研究领域为数据挖掘,联机分析处理,模式识别;马秀莉(1972?),女,博士后,主要研究领域为数据挖掘,联机分析处理;谭少华(1960?),男,博士,教授,博士生导师,主要研究领域为人工智能,模式识别; 唐世渭(1939?),男,教授,博士生导师,CCF高级会员,主要研究领域为数据库,信息系统;杨冬青(1945?),女,教授,博士生导师,CCF高级会员,主要研究领域为数据库,信息系统. 遇 辉1,2+, 马秀莉1,2, 谭少华1,2, 唐世渭1,2, 杨冬青1 1(北京大学 信息科学技术学院,北京 100871) 2(北京大学 视觉与听觉信息处理国家重点实验室,北京 100871) Exceptional Slices Mining Based on Singular Value Decomposition YU Hui1,2+, MA Xiu-Li1,2, TAN Shao-Hua1,2, TANG Shi-Wei1,2, YANG Dong-Qing1 1(School of Electronics Engineering and Computer Science, Peking University, Beijing 100871, China) 2(National Laboratory on Machine Perception, Peking University, Beijing 100871, China) + Corresponding author: Phn: +86-10 Fax: +86-10 E-mail: yuhui@, Received 2004-07-16; Accepted 2005-03-11 Yu H, Ma XL, Tan SH, Tang SW, Yang DQ. Exceptional slices mining based on singular value decomposition. Journal of Software, 2005,16(7):1282?1288. DOI: 10.1360/jos161282 Abstract: Slice is one of the major operations in on-line analysis processing, which has played an important role in the application of decision support. In this paper, a method of mining exceptional slices is presented for extracting the distribution feature of the slice data based on the technique of the singular value decomposition, and the exceptional slices can be found by utilizing the distance-based outlier detection technique on the singular value feature. The effectiveness of the approach is experimentally demonstrated on the artificial data and the real slices data. Key words: exceptional slices mining; feature extraction; singular value feature vector; distance-based outlier detection; on-line analytical processing 摘 要: 切片操作是联机分析处理