基于扩展关系模型的多Bayesian网依赖结构的合并  

Integrating Multi Dependency Structure of Bayesian Network Based on Generalized Relation Model

作　　者：李维华[1] 刘惟一[1] 张忠玉[1] 郭祥文[1] 张燕峰[1] 

机构地区：[1]云南大学计算机科学系,昆明650091

出　　处：《计算机科学》2004年第7期192-195,共4页Computer Science

基　　金：国家自然科学基金:No.60263006;中国科学院计算机智能信息处理协会重点实验室项目:No.Ⅱ p2002-2;云南自然科学基金:No.2002F0063M

摘　　要：贝叶斯网是一个每个结点都带有一张概率表的有向无环图,它可以有效地表示不确定性知识并进行知识推理。知识系统在很多时候不得不将来自不同信息源或者同一信息源不同时刻的知识合并起来。Bayesian网作为一个知识系统,所以也会面临将多个Bayesian网提供的信息结合起来。本文提出一个基于扩展的关系数据模型和条件独立的算法,该算法将多个Bayesian网合并成为一个Bayesian网,并且尽可能地保留每一个Bayesian网的信息。A Bayesian Network is a directed acyclic graph (DAG) with conditional probabilities for each node. Bayesian network is a powerful common knowledge representation and reasoning tool for partial beliefs under uncertainty. However, Knowledge-based system sometimes must be able to 'intelligently' manage a large amount of information coming from different sources and at different moments in times. Based on generalized relation and the conditional independences defined by the Bayesian Network we present an algorithm that integrates multi Bayesian network and construct a large Bayesian Network preserving as much information as possible.

关 键 词：扩展关系模型 多Bayesian网 贝叶斯网 | 条件独立 多值依赖 

分 类 号：TP311.12[自动化与计算机技术—计算机软件与理论]