基于贝叶斯网络的数据挖掘方法  

Data Mining Based on Bayesian Networks

作　　者：李艳美[1] 张卓奎[1] 

机构地区：[1]西安电子科技大学理学院,陕西西安710071

出　　处：《计算机仿真》2008年第2期87-89,161,共4页Computer Simulation

基　　金：国家自然科学基金资助课题(70371031);国家自然科学基金资助课题(70373046)

摘　　要：常用的数据挖掘方法有许多,贝叶斯网络(Bayesian Networks,BN)方法在数据挖掘中的应用是当前研究的热点问题。贝叶斯网络是一种进行不确定性推理和知识表示的有力工具,当与统计方法结合使用时,显示出许多关于数据处理的优势。首先介绍了BN的定义、方法的优点以及目前网络学习的各种算法,最后用一个实际中的案例进行试验,指出了在数据挖掘技术中的具体应用。得到了将贝叶斯网络应用于数据挖掘当中,充分挖掘数据的隐含信息和内在本质,具备良好地预测能力等优点,实验证明这种方法实用、有效。Many methods have been used in data mining, Bayesian networks become a focus currently. It is a powerful tool and can be used to do uncertain inference. When used in conjunction with statistical techniques, Bayesian networks have several advantages for data modeling. This paper mainly discusses the definition of Bayesian net-works, the advantages of the methods and the algorithms of learning networks. An instance is presented to indicate the applications in data mining technology finally. Bayesian networks can mine the hidden information and inherent hypostasis, and have many advantages such as the ability of favorable forecast. The experimentation proves its practi- cality and availability.

关 键 词：贝叶斯网络 数据挖掘 条件概率分布 条件独立性 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程] TP301[自动化与计算机技术—控制科学与工程]