摘要
通过对ＫＤＤ中基于相联规则的经典数据挖掘算法的讨论，提出构造测试平台的思想，旨在借助测试指标进一步指导和规范数据挖掘算法。在采用测试数据方面，阐明基于测试指标的人造数据比模拟现实世界的人造数据更适合于做为测试平台的数据集的思想。指出对Ａｐｒｉｏｒｉ等经典算法的改进途径。由于二次挖掘可以用前次挖掘出的知识作指导，从而提高了数据挖掘效率，因此，二次挖掘算法将成为今后ＫＤＤ领域中研究的热点。
In this paper the algorithms of data mining based on association rules are discussed. An environment of algorithms tested is very important. This environment can lead the studies of algorithms on data mining and help somebody determine choice one to use. The synthetic data problem-oriented is more useful than the synthetic data stimulation - oriented in the test environment. We have advanced some ideas of data mining after studied Apriori. The second mining will improve efficiency of KDD for lots of rules which had been mined can be used.
出处
《吉林工业大学自然科学学报》
CSCD
2000年第2期43-46,共4页
Natural Science Journal of Jilin University of Technology
基金
国家自然科学基金资助项目!(69873019)
