A NEW METHOD TO MINING GLOBAL PECULIAR DATA BASED ON CLUSTERING

摘要：

Mining peculiar data is an important task of data mining. Data are peculiar if they are rare and are very different from others. Concepts and methods of two present peculiar data mining frames are introduced. A method to mining global peculiar data based on clustering is proposed. Character of peculiar data is utilized adequately in the method. Computing for data that are impossibly peculiar is avoided mostly in the method. Experiments show good performance of the method.

关键词： data mining peculiar data clustering global

作者: Li Yingjie Yin Yixin

作者单位: Information Engineering Department,University of Science and Technology Beijiing,Beijing 100083,Chin Information Engineering Department,University of Science and Technology Beijiing,Beijing 100083,Chin

会议类型: 国际会议

会议名称: 2008年拟人系统国际会议(2008 International Conference on Humanized Systems )(ICHS’08)

会议地点: 北京

会议语种:英文

在线出版日期: 2008-10-18（万方平台首次上网日期，不代表论文的发表时间）