基于命题逻辑的关联规则挖掘算法L-Eclat

发布时间：2018-05-05 13:20

本文选题：关联规则 + 命题逻辑　；参考：《计算机科学》2017年12期

【摘要】：关联规则挖掘是数据挖掘领域非常重要的课题,在很多领域被广泛应用。关联规则挖掘算法都需要设置最小支持度和最小置信度。很多国内外学者研究的挖掘算法在这两方面都存在着一些问题,不仅需要大量的领域知识来设置合适的最小支持度,而且其结果集庞大、用户不容易理解。针对关联规则挖掘算法存在的问题,将命题逻辑融合到关联规则算法Eclat中,设计出了基于命题逻辑思想的挖掘算法L-Eclat。实验结果表明,L-Eclat算法压缩了挖掘的规则集,减小了算法的时间消耗,且即使是非常小的支持度也可以得到高质量的关联规则,这在一定程度上解决了支持度设置的问题。
[Abstract]:Association rules mining is a very important topic in the field of data mining. It is widely used in many fields. Association rules mining algorithms need to set minimum support and minimum confidence. Many domestic and foreign scholars research algorithms have some problems in these two aspects, not only need a large number of domain knowledge to set the appropriate. Minimum support degree, and its result set is huge, the user is not easy to understand. Aiming at the problem of association rule mining algorithm, the thesis combines propositional logic to association rule algorithm Eclat, and designs a mining algorithm based on propositional logic L-Eclat. experiment results. The L-Eclat algorithm compresses the rule set of mining and reduces the algorithm. Time consuming and even very small support can get high quality association rules, which solves the problem of support setting to a certain extent.

【作者单位】：浙江工业大学计算机科学与技术学院;
【基金】：浙江省自然科学基金(LY14F020018)资助
【分类号】：TP311.13

【相似文献】