节点文献
基于排序树的频繁项集挖掘算法
Frequent itemsets mining algorithm based on sort tree
【摘要】 提出了排序树的概念以及用排序树存储频繁项集的思想,证明了末项剪枝性质,以O(1)的时间实现了与Apriori算法同样的连接和剪枝操作,采用祖先兄弟表示法存储排序树,在某事务不包含祖先时跳过具有共同祖先的所有兄弟结点,提高了计算支持度的时间性能。理论分析和实验结果均表明,在时间性能方面本文算法与Apriori算法相比有较大提高。
【Abstract】 In this paper,the concept of sort tree is proposed,and the sort tree is used to store frequent itemsets,which proves the property of the last item pruning.Joining and pruning operations a re implemented as Apriori algorithm with O(1)time complexity.Ancestor-brother express is used to store the sort tree.If an ancestor does not exist in a transaction,all the brother nodes with the same ancestor will be skipped.So the time performance of counting support can be improved.Theoretical analysis and experimental results show that the proposed algorithm can improve the time performance greatly compared with Apriori algorithm.
【Key words】 artifical intelligence; frequent itemsets; last term pruning; sort tree; ancestor-brother express;
- 【文献出处】 吉林大学学报(工学版) ,Journal of Jilin University(Engineering and Technology Edition) , 编辑部邮箱 ,2016年04期
- 【分类号】TP311.13
- 【被引频次】6
- 【下载频次】115