摘要
Data mining is the process of discovering hidden structure or patterns in large quantities of data by usingkinds of analytic tools. The structure or patterns can help decision makers for advantageous actions. This paper intro-duces the concept of interestingness and reference rules ,and uses interestingness to estimate the information includedin rule,and then presents a method for mining exception rules while computing the interestingness according to thereference rules. Experiments compared with other methods show that the proposed method has the better effects.
Data mining is the process of discovering hidden structure or patterns in large quantities of data by using kinds of analytic tools. The structure or patterns can help decision makers for advantageous actions. This paper introduces the concept of interestingness and reference rules,and uses interestingness to estimate the information included in rule,and then presents a method for mining exception rules while computing the interestingness according to the reference rules. Experiments compared with other methods show that the proposed method has the better effects.
出处
《计算机科学》
CSCD
北大核心
2003年第3期40-43,共4页
Computer Science
基金
国家自然科学基金(69733030)
广东省自然科学基金(001264)
广东省教育厅<软件技术>重点实验室研究基金
