【摘要】 提出了一种适用于证券业交易数据库挖掘的频繁模式链表关联规则挖掘算法,它采用共享前缀交易项树和频繁模式链表结构,无须产生候选项集,FPL-growth算法通过直接排列出链路中的频项组合,就可得到完整的频繁模式集,且支持多阈值挖掘,挖掘到的关联规则带有时间段属性,特别适用于证券业的个性化信息需求获取。更多还原

【Abstract】 This paper proposes an association rule algorithm which is the same with trade database of stock.It uses share prefix transaction tree and frequent pattern link(FP-link) data structure,and allows frequent pattern mining without generation of candidate itemsets.FPL-growth algorithm gains frequent pattern sets via straight arrange combination of frequent itemsets within link.It can be used for multiple frequent pattern mining with different supports and obtained association rule has property of time,especially applies in stockjobber to gain personal information requirement.更多还原