基于加权有向图的权频繁模式挖掘算法  

Mining weighted frequent patterns algorithm based on weighted graph

作　　者：封军[1] 郑诚[1] 郑小波[1] 肖云[1] 

机构地区：[1]安徽大学计算智能与信号处理教育部重点实验室,安徽合肥230039

出　　处：《微型机与应用》2010年第20期4-7,共4页Microcomputer & Its Applications

摘　　要：为解决加权图遍历模式的挖掘问题,提出了一种从加权有向图中挖掘加权频繁模式算法。在该算法中,利用图全局拓扑结构和顶点权值信息评估遍历模式的权支持度,从而将剪枝问题转化成模式可扩展性问题,再利用可扩展模式产生候选模式集。本算法把图顶点权值融合进来,提高了挖掘结果的准确度。实验结果表明,该算法可以有效地进行基于加权向图的权频繁模式挖掘。In order to better solve the problem about mining weighted frequent patterns based on the weighted graph,the related algorithm is proposed,which is to mine weighted frequent patterns form graph with weight on its vertexes. In this algorithm,we use global topological structure and information to assess the weight support of traversal patterns,and then pruned problem is transformed into patterns scalability problems,and to utilize the extensible patterns to produce the candidate patterns. This algorithm mainly combines with the weight of vertexes to enhance the accuracy of the results. The experimental result shows that it is efficient to mine weighted frequent patterns from the weighted graph.

关 键 词：加权有向图 权支持度 关联规则 权频繁模式 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]