基于FP-tree挖掘密集型数据最大频繁模式算法

An Algorithm for Mining Maximal Frequent Patterns of Dense Datasets Based on Fp-tree

Fp-growth算法是当前挖掘频繁项目集算法中速度最快,应用最广,并且不需要候选集的一种挖掘关联规则的算法．但是,Fp-growth算法也存在着算法结构复杂和空间利用率低等缺点．在FP-tree结构的基础上提出了密集型数据最大频繁模式挖掘算法FP-DMax．算法FP-DMax只需要2次扫描数据库,在挖掘过程中不产生候选项集,大大提高了算法的时空效率．实验表明,算法FP-DMax在挖掘密集型数据最大频繁模式方面是高效的．

Fp-growth algorithm is one of the currently fastest and most popular one for mining association rule without candidate generation. However, it has disadvantages such as complicated data structure and lower space utilization rate. This paper develops the algorithm FP-DMax for mining maximal frequent patterns of dense datasets based on the data structure FP-tree. The algorithm only scans the database twice and generates no candidate itemsets. The experiment shows that the algorithm FP-DMax is efficient on mining dense datasets.