【摘要】 超团模式是一种新型的关联模式,这种模式所包含的项目相互间具有很高的亲密度.超团模式中某个项目在事务中的出现很强地暗示了模式中其他项目也会相应地出现.极大超团模式是一组超团模式更加紧凑的表示,可被用于多种应用.挖掘这两种模式的标准算法是完全不同的.提出一种基于FP-tree(frequent pattern tree)的快速挖掘算法——混合超团模式增长(hybrid hyperclique pattern growth,简称HHCP-growth),统一了两种模式的挖掘.算法采用递归挖掘方法,并应用多种有效的剪枝策略.提出并证明几个相关命题来说明剪枝策略的有效性和算法的正确性.实验结果表明,HHCP-growth算法相对于标准的超团模式挖掘算法和极大超团模式挖掘算法都具有更高的效率,尤其对于大数据集或在低支持度条件下更为显著.更多还原

【Abstract】 The hyperclique pattern is a new type of association pattern, in which items are highly affiliated with each other. The presence of an item in one transaction strongly implies the presence of every other item in the same hyperclique pattern. The maximal hyperclique pattern is a more compact representation of a group of hyperclique patterns, which is desirable for many applications. The standard algorithms mining the two kinds of patterns are different. This paper presents a fast algorithm called hybrid hyperclique pattern growth (HHCP-growth) based on FP-tree (frequent pattern tree), which unifies the mining processes of the two patterns. This algorithm adopts recursive mining method and exploits many efficient pruning strategies. Some propositions are also presented and proved to indicate the effectiveness of the strategies and the validity of the algorithm. The experimental results show that HHCP-growth is more effective than the standard hyperclique pattern and maximal hyperclique pattern mining algorithms, particularly for large-scale datasets or at low levels of support.更多还原