一种挖掘free项目集的快速算法  

An Algorithm for Fast Free- Sets Mining

作　　者：赵栋[1] 卢炎生[1] 王涛[1] 

机构地区：[1]华中科技大学计算机学院,湖北武汉430074

出　　处：《小型微型计算机系统》2004年第10期1853-1856,共4页Journal of Chinese Computer Systems

摘　　要：对关联规则的挖掘是数据挖掘中一个重要的问题 .通过挖掘 free项目集来挖掘关联规则已被证明是一种十分高效的方法 .Seg Free算法将数据库分成许多分段并在这些分段中查找 free项目集 .它只耗用很小的额外内存来存储在每个分段中项目集的支持度 ,却能极大的减少项目集匹配的时间 ,而项目集匹配的时间是整个挖掘过程的瓶颈 .在真实数据集上的试验已显示了它良好的性能 .Mining of association rule is an important problem in data mining. Ming of free sets has proved to be an efficient way for association rule mining. The SegFree algorithm divide the database into segments and find free sets in each segment. The SegFree algorithm costs little memory to save additional support number of itemsets in each segment but greatly reduced the time of itemset matching which is the bottleneck of the mining process. The experiments on real datasets have showed its good performance. It can also be used in some other mining tasks.

关 键 词：数段挖掘 分段 关联规则 free项目集 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]