基于WNegNodeset结构的加权频繁项集挖掘算法  

Mining frequent weighted itemsets based on WNegNodeset structure

作　　者：王斌 房新秀 吕瑞瑞 马俊杰 Wang Bin;Fang Xinxiu;Lyu Ruirui;Ma Junjie(School of Information&Control Engineering,Qingdao Technological University,Qingdao Shandong 266520,China)

机构地区：[1]青岛理工大学信息与控制工程学院,山东青岛266520

出　　处：《计算机应用研究》2020年第7期1989-1992,2010,共5页Application Research of Computers

基　　金：国家自然科学基金资助项目(61502262)。

摘　　要：针对基于WN-list加权频繁项集挖掘算法(NFWI)中挖掘加权频繁项集(FWI)效率低的问题,提出了一种基于WNegNodeset结构的加权频繁项集挖掘算法(NegNFWI)。该算法首先采用了新的数据结构WNegNodeset,它是NegNodeset的扩展,该数据结构采用了一种新的基于集合位图表示的位图加权树(BMW-tree)节点编码模型,通过按位运算符快速提取WNegNodeset的节点集,避免了大量的交集运算;其次采用了差集策略快速计算项集的加权支持度,从而减少了计算量;最后通过仿真实验验证了算法的有效性和可行性。In the mining algorithm for frequent weighted itemsets based on WN-list(NFWI),mining weighted frequent itemsets(FWI)is inefficient.To solve the problem,this paper proposed a frequent weighted itemsets mining algorithm(NegNFWI)based on WNegNodeset structure.Firstly,this algorithm used the data structure of WNegNodeset,an extension of NegNodeset.The data structure employed a novel encoding model for nodes in bitmap weighted-tree(BMW-tree)based on the bitmap representation of sets,and used bitwise operators to extract WNegNodesets of itemsets quickly,avoiding a large quantity of intersection operations.Secondly,this algorithm used diffsets strategy to calculate the weighted support degree of itemsets quickly,thus decreasing computing time.Finally,results from simulation experiments show that the proposed algorithm is efficient and feasible.

关 键 词：加权频繁项集 加权支持度 位图加权树 按位运算符 差集策略 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]