| 摘要: |
| 发现最大频繁项目集是多种数据开采应用中的关键问题，提出一种快速开采最大频繁项目集的算法DMFI(discovery maximum frequent itemsets), 该算法把自底向上和自顶向下的搜索策略进行了合并。通过其独特的排序方法和有效的剪枝策略，大大减少了修选项目集的生成，从而显著地降低了CUP时间。 |
| 关键词: 数据开采 知识发现 关联规则 最大频繁项目集 最小支持度 |
| DOI： |
| 分类号: |
| 基金项目:国家“九五”国防预研基金资助项目。 |
|
| Fast Mining Maximum Frequent Itemsets |
|
Lu Song-feng,Lu Zheng-ding
|
| Abstract: |
| Discovering maximum frequent itemsets is a key problem in many data mining applications. In this paper,the DMFI (discovery maximum frequent itemsets) algorithm which combines the bottom-up and top-down searches is proposed to solve this problem. Using the unique ordering method and efficient pruning strategy, the number of candidete itemsets is greatly decrcased, therefore CPU time is reduced remarkably. |
| Key words: data mining knowledge discovery association rule maximum frequent itemset: minimum support |
