基于贝叶斯扩张树的文本分类算法  

Algorithm of documents categorization based on Bayesian-spanning tree

作　　者：史会峰[1] 谷根代[1] 卢艳霞[1] 

机构地区：[1]华北电力大学数理学院,河北保定071003

出　　处：《华北电力大学学报（自然科学版）》2004年第5期105-107,共3页Journal of North China Electric Power University：Natural Science Edition

基　　金：华北电力大学青年基金资助项目(060203).

摘　　要：将文本的特征作为贝叶斯网络的节点,根据样本数据计算节点之间的互信息,并且将它们作为网络中边的权。利用 Kruskal 算法构造了一种简单的贝叶斯网络—贝叶斯扩张树,使它具有最大对数似然值。利用贝叶斯条件概率公式,将贝叶斯扩张树作为分类器对文档集中的文档进行分类。Taking the characteristics of a document as the nodes in network,the weight of an edge is the mutual information between two nodes.Using the Kruskalthe algorthim,a simple Bayesian network-Bayesian spanning tree with the maximum log-liklihood is constructed.By the conditional probability formula,regarding the Bayesian spanning tree as an classifier,the documents in the documents set are classified.The experimental results indicate that this approach is more effective and has higher accuracy than the naive Bayesian classifier.

关 键 词：文本分类 贝叶斯网络 分类器 文档 KRUSKAL算法 互信息 节点 扩张 样本数据 最大 

分 类 号：TM732[电气工程—电力系统及自动化] TP391[自动化与计算机技术—计算机应用技术]