摘要
挖掘频繁项集是许多数据挖掘任务中的关键问题,也是关联规则挖掘算法,所以提高频繁项集的生成效率一直是近几年数据挖掘领域研究的热点之一,研究人员从不同的角度对算法进改进以提高算法的效率。该文提出了一种基于位表的频繁项集挖掘算法,用一种特别的数据结构———位表来压缩数据库以便快速产生候选集和支持计数,实验结果表明;此算法大大减少了遍历的时间,是性能比较好的算法。
Mining the frequent iternsets is a key problem in data mining. It is also the core of the algorithm for mining association rules. Therefore, improving the efficiency of discovering the algorithms from different perspectives has been the study focus. In the paper, an effective algorithm named as Bit- TableFI was presented. In the algorithms, a special data structure BitTable was used to compress database for quick candidate itemsets generation and support count. Experiment shows that the algorithm outperforms Apriori.
出处
《贵州工业大学学报（自然科学版）》
CAS
2006年第6期60-63,69,共5页
Journal of Guizhou University of Technology(Natural Science Edition)
关键词
数据挖掘
频繁项集
位表
data mining
frequent itemsets
bittable
