基于时间衰减模型的模糊会话关联规则挖掘算法  

Fuzzy session association rule mining algorithm based on time decay model

作　　者：杨英杰[1,2] 邱卫[1,2] 

机构地区：[1]解放军信息工程大学,郑州450001 [2]河南省信息安全重点实验室,郑州450001

出　　处：《计算机应用研究》2017年第1期128-131,140,共5页Application Research of Computers

基　　金：国家"863"计划资助项目(2012AA012704);国家"973"计划资助项目(2011CB311801);郑州市科技领军人才项目(131PLJRC644)

摘　　要：现有的关联规则挖掘算法没有考虑数据流中会话的非均匀分布特性和历史数据的作用,并且忽略了连续属性处理时的尖锐边界问题。针对这些问题,提出一种基于时间衰减模型的模糊会话关联规则挖掘算法。针对数据流中会话的非均匀分布特性,基于时间片对会话进行划分,完整地保留了时间片内会话之间的相关性信息,采用模糊集对会话的连续属性进行处理,增加了规则的兴趣度和可理解性。在考虑历史数据作用和允许误差情况的基础上,基于时间衰减模型挖掘数据流中的临界频繁项集和模糊关联规则。实验结果表明,该方法在提高时间效率、降低冗余率和增加规则兴趣度方面存在明显优势。The existing algorithms do not consider the non-uniform distribution feature of session and the effect of historical data in association rule mining. It also cannot handle the sharp boundaries problem of continuous attributes. As to solve these problems, this paper proposed a fuzzy session association rule mining algorithm based on time decay model. First, for the nonuniform distribution feature of data stream session, it used the time slice to divide the data stream for retaining the relevant relationship information of sessions. Then, applied the fuzzy set to discrete continuous attributes, increasing the interestingness and comprehensibility of the rules. Finally, it used the time decay factor and error factor in fuzzy association rules mining. And experimental results show that this method has obvious advantages in improving time efficiency, reducing redundancy and increasing the interest of rules.

关 键 词：数据流 时间片 关联规则 模糊集 衰减模型 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术] TP301.6[自动化与计算机技术—计算机科学与技术]