关联规则挖掘AprioriTid算法优化研究

Study on Optimization of AprioriTid Algorithm for Mining Association Rules

提出了一种基于事务压缩和项目压缩的AprioriTid优化算法。该算法的特点是:项目集采用关键字识别,同时对事务数据进行事务和项目压缩。从而省去了Apriori算法和AprioriTid算法中的剪枝和模式匹配步骤,减小了扫描事务数据库的大小,提高了发现规则的效率。通过实验表明,优化的算法执行效率明显优于AprioriTid算法。

This paper puts forward an optimizied algorithm which associates AprioriTid with transaction reduction and item reduction technique. Its characteristic is that the candidate set is adopted by the key word identifies, and at the same time transaction data is compressed by transaction and item. So the process of pruning and string pattern matching in AprioriTid and Apriori algorithm are removed, the size of scan transaction data base is decreased, and efficiency of find rules is improved. The testing result shows that the performance efficiency of optimized algorithm is obviously better than AprioriTid algorithm.