Algorithm for Association Rule Mining Based on Sorting Matrix and Tree
Generating the frequent itemsets is a key problem of association rule mining. In Apriori algorithm, the procedure which determines the frequent itemsets from a huge number of candicate itemsets is extremely time-consuming. This paper proposes a new algorithm, which combines the ordinal character of itemsets to create matrix and sorts the nodes in the order of ascending support count in the tree, so that the sets of the candidate frequent itemsets is the smallest totally. In this algorithm, the number of the candidate frequent itemsets can be greatly reduced and the completness of frequent itemsets are kept, so that the cost of computing is reduced and the efficiency of algorithm is improved.
Association Rule Apriori Algorithm Itemsets Ordered Frequent Itemsets
DUAN Longzhen ZHU Yixia HUANG Longjun HUANG Shuiyuan
Department of Computer Science Nanchang University Nanchang, JiangXi Province, China Software School Jiangxi Normal University Nanchang, Jiangxi Province, China
国际会议
厦门
英文
755-758
2006-07-27（万方平台首次上网日期，不代表论文的发表时间）