基于广义离散Morse理论的强关联规则挖掘  

Strong-association Rules Mining Based on Generalized Discrete Morse Theory

作　　者：刘俊[1] 刘希玉[1] 

机构地区：[1]山东师范大学管理与经济学院,济南250014

出　　处：《计算机工程》2011年第16期45-47,共3页Computer Engineering

基　　金：国家自然科学基金资助项目(60873058);山东省自然科学基金资助项目(Z2007G03)

摘　　要：针对强关联规则的挖掘问题,提出构造事务数据库的单元复形,利用广义离散Morse理论发现强关联规则的方法。在基本的离散Morse理论和关联规则的基础上延伸得到广义离散Morse理论和强关联规则的定义,通过在事务数据库的单元复形上定义离散Morse函数挖掘强关联规则,例证表明该方法的可行性和高效性。For the problem of strong-association rules mining,a method is proposed which constructs a cell complex on transaction database and uses generalized discrete Morse theory to find the strong-association rule.It gets the definition of generalized discrete Morse theory and strong-association rule by extending the basic discrete Morse theory and association rule,mining the strong-association rule by defining discrete Morse theory on cell complex of transaction database.Example verifies the feasibility and efficiency of the method.

关 键 词：离散Morse理论 拓扑 广义离散Morse函数 广义离散梯度 强关联规则 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]