基于抽样的分布式约束性关联规则挖掘算法研究

本文采用抽样的方法，在基于约束的Eclat类算法（例如Eclat A和Eclat M）的基础上，提出了一种分布式约束性关联规则的挖掘算法——DMCASE算法。本算法在各数据站点上对一个较小的样本采用基于约束的Eclat类算法，挖掘局部约束频繁项集，采用归纳学习的方法归并所有局部约束频繁项集，产生全局约束频繁项集。只需1次扫描数据库，挖掘效率较高。实验证明：该算法是一种十分有效的解决基于约束条件下的分布式关联规则挖掘算法。

Algorithm Analysis for Anti-monotone and Monotone Constraints Association Rules Mining

An algorithm for distributed mining association rules with constraints, called DMCASE, is presented using sampling and constrained Eclat algorithm. At each database sites, sampling algorithm and constrained Eclat algorithm are implemented. And the local frequent itemsets that satisfied constraints are developed. They then are combined to global frequent itemsets that satisfied constraints based on learning from induction. DMCASE algorithm scans the whole database only once. It also is an algorithm with high efficiency. Results from our experiments show that the al- gorithm is an effective way to resolve the problem of distributed mining association rules with constraints.