基于频繁项集树的时态关联规则挖掘算法

Temporal association rules mining algorithm based on frequent item sets tree

针对目前时态关联规则研究中存在的挖掘效率不高、规则可解释性低、未考虑项集时间关联关系等问题,在原有相关研究的基础上,提出一种新的基于频繁项集树的时态关联规则挖掘算法.通过对时间序列数据进行降维离散化处理,采用向量运算生成频繁项集,提高频繁项集挖掘效率.考虑到项集之间的时态关系以及树结构的优势,提出一种新的频繁项集树结构挖掘时态关联规则,其挖掘频繁项集与树结构构建同时进行,无需产生候选项集,提高了规则挖掘效率.实验表明,对比于其他算法,所提出算法在挖掘效率和规则解释性方面效果更好,具有较好的应用前景.

In order to improve the efficiency and enhance the interpretability in mining the temporal association rules,a new mining algorithm of temporal association rules based on frequent iemsets tree is proposed. The time series data is discretized after the dimension reduction, on this basis, vector operations are adopted to generate frequent itemsets to improve the efficiency. In view of the advantage of the structure of the tree and the temporal interval relation between the items, a frequent itemsets tree is constructed in parallel with mining frequent itemsets to improve the efficiency of rule mining without generating candidate itemsets. Experimental results show that the proposed algorithm can provide better efficiency and interpretability in mining temporal rules in comparison with other algorithms and has good application prospects.