基于十字链表的关联规则改进算法  

Improved algorithm of association rules based on orthogonal list

作　　者：王利钢[1] 陈平[1] WANG Li-gang CHEN Ping(Nanjing College of Information Technology, Nanjing 210046, Jiangsu, China)

机构地区：[1]南京信息职业技术学院,江苏南京210046

出　　处：《宝鸡文理学院学报（自然科学版）》2016年第4期27-30,共4页Journal of Baoji University of Arts and Sciences(Natural Science Edition)

基　　金：江苏高校哲学社会科学研究课题(2015SJB311);2014年南京信息职业技术学院"青蓝工程"资助

摘　　要：目的解决在关联规则挖掘中存在大量无用的候选项集和频繁项集以及效率不高的问题。方法提出了一种基于十字链表的关联规则改进算法。结果利用用户给定的条件和剪枝算法减少了候选项集的数量,再遍历十字链表确定频繁项集,结果表明算法提高了挖掘效率。结论基于十字链表的关联规则改进算法避免了重复扫描数据库,提高了挖掘效率。Purposes--To deal with the problems about lots of useless candidate sets and frequent sets as well as low efficiency when mining association rules. Methods--An improved algorithm of association rules based on orthogonal list is proposed. Results--The number of candidate sets is reduced by making use of the conditions given by the users and pruning algorithm. Then frequent sets are generated by traversing the orthogonal list. Results show that the algorithm improves the mining efficiency. Conclusions--An improved algorithm of association rules which is based on orthogonal list can avoid scanning database repeatedly and improves the mining efficiency.

关 键 词：关联规则 十字链表 数据挖掘 频繁项集 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]