一种基于分布式数据库的全局频繁项目集更新算法

Algorithm based on distributed database for updating global frequent itemsets

在算法FMAGF的基础上 ,提出了一种基于分布式数据库的全局频繁项目集更新算法———UAGFI,该算法主要考虑最小支持度发生变化时全局频繁项目集的更新情况 .UAGFI在最坏的情况下仅须扫描各局部数据库一遍 ,并利用已挖掘的结果 ,可避免传送某些原全局频繁项目对应的条件频繁模式树 ,从而降低网络通讯代价 .实验结果表明 。

A new algorithm UAGFI (updating algorithm of global frequent itemsets based on distributed database) is introduced, it considers the updating of global frequent itemsets when dynamically adjusting minimum support measure threshold. In the worst case, UAGFI only scans every local transaction database once and can avoid transmitting some conditional pattern tree and/or base of original global frequent item by utilizing mined results. Therefore, UAGFI uses far less communication overhead and obviously improves updating efficiency of global frequent itemsets. Experimental results show that UAGFI algorithm is efficient and effective.