利用matlab进行简单的贝叶斯网络构建

matlab的安装

第一次装matlab，装好后发现没有在桌面生成图标，还以为是最后提示的编译器没有安装。结果发现，matlab的图标在其安装的bin目录下面，进行初始化就能开始使用了

关于贝叶斯网络

1.简单的教程可以参照如下地址：

http://cs.nyu.edu/faculty/davise/ai/bayesnet.html
http://www.cnblogs.com/leoo2sk/archive/2010/09/18/bayes-network.html

2.向matlab中添加FULLBNT，参考地址：

http://hi.baidu.com/73290673/item/21db99f36d90bc49932af29d

采用MATLAB语言编制的贝叶斯网络工具箱(Bayesian Networks Toolbox,BNT)可实现贝叶斯网络结构学习、参数学习、推理和构建贝叶斯分类器,此工具箱在贝叶斯学习编 程方面非常灵活。

官方主页：http://www.cs.ubc.ca/~murphyk/Software/BNT/bnt.html

官方下载：http://www.cs.ubc.ca/~murphyk/Software/BNT/FullBNT-1.0.4.zip

原文链接：http://hi.baidu.com/zgyz/blog/item/2d3627f415c7fbe77709d763.html

贝叶斯网络：http://www.cs.ubc.ca/~murphyk/Software/BNT/bnt.html

语音工具箱：http://www.ee.ic.ac.uk/hp/staff/dmb/voicebox/voicebox.html

1、解压FullBNT-1.0.4.zip，将整个目录FullBNT-1.0.4复制到MATLAB的安装目录的TOOLBOX目录下，如D:\MATLAB7\toolbox\

2、打开Matlab，在MATLAB命令窗口中输入以下命令：

>> cd D:\MATLAB7\toolbox\FullBNT-1.0.4

>> addpath(genpathKPM(pwd))

>> 将TOOLBOX下新加的BNT工具箱加到MATLAB的搜索路径中去。 添加BNT工具箱的MATLAB的搜索路径也可采用如下指令

>> addpath(genpath('D:\MATLAB7\toolbox\FullBNT-1.0.4'))

>>

3、为了永久保存上面的路径，以免下次重启MATLAB时重新添加，在MATLAB命令窗口下使用下面的命令：

>> savepath >>

4、检验是否成功设置的方法： 在命令窗口中输入以下命令：which test_BNT.m（可以为所加工具箱的任一个M文件名称），如果显示正确，就说明上面的设置成功。

>> which test_BNT.m D:\MATLAB7\toolbox\FullBNT-1.0.4\BNT\test_BNT.m

>>

3.关于FULLBNT使用简单教程

http://www.cs.ubc.ca/~murphyk/Software/BNT/usage.html#examples

http://blog.sina.com.cn/s/blog_6c7b434d01013ufz.html

实例分析

如下图所示，给出了贝叶斯网络的图和相应的条件概率



对上述信息建立贝叶斯网络，代码如下

[plain] view plaincopyprint?

N=8;

dag=zeros(N,N);

A=1;S=2;T=3;L=4;B=5;E=6;X=7;D=8;

dag(A,T)=1;

dag(S,[L B])=1;

dag([T L],E)=1;

dag(B,D)=1;

dag(E,[X D])=1;

discrete_nodes=1:N;

node_sizes=2*ones(1,N);

bnet=mk_bnet(dag,node_sizes,'names',{'A','S','T','L','B','E','X','D'},'discrete',discrete_nodes);

bnet.CPD{A}=tabular_CPD(bnet,A,[0.99,0.01]);

bnet.CPD{S}=tabular_CPD(bnet,S,[0.5,0.5]);

bnet.CPD{T}=tabular_CPD(bnet,T,[0.99,0.95,0.01,0.05]);

bnet.CPD{L}=tabular_CPD(bnet,L,[0.99,0.9,0.01,0.1]);

bnet.CPD{B}=tabular_CPD(bnet,B,[0.7,0.4,0.3,0.6]);

bnet.CPD{E}=tabular_CPD(bnet,E,[1,0,0,0,0,1,1,1]);

bnet.CPD{X}=tabular_CPD(bnet,X,[0.95,0.02,0.05,0.98]);

bnet.CPD{D}=tabular_CPD(bnet,D,[0.9,0.2,0.3,0.1,0.1,0.8,0.7,0.9]);

draw_graph(dag)

N=8;
dag=zeros(N,N);
A=1;S=2;T=3;L=4;B=5;E=6;X=7;D=8;
dag(A,T)=1;
dag(S,[L B])=1;
dag([T L],E)=1;
dag(B,D)=1;
dag(E,[X D])=1;
discrete_nodes=1:N;
node_sizes=2*ones(1,N);
bnet=mk_bnet(dag,node_sizes,'names',{'A','S','T','L','B','E','X','D'},'discrete',discrete_nodes);
bnet.CPD{A}=tabular_CPD(bnet,A,[0.99,0.01]);
bnet.CPD{S}=tabular_CPD(bnet,S,[0.5,0.5]);
bnet.CPD{T}=tabular_CPD(bnet,T,[0.99,0.95,0.01,0.05]);
bnet.CPD{L}=tabular_CPD(bnet,L,[0.99,0.9,0.01,0.1]);
bnet.CPD{B}=tabular_CPD(bnet,B,[0.7,0.4,0.3,0.6]);
bnet.CPD{E}=tabular_CPD(bnet,E,[1,0,0,0,0,1,1,1]);
bnet.CPD{X}=tabular_CPD(bnet,X,[0.95,0.02,0.05,0.98]);
bnet.CPD{D}=tabular_CPD(bnet,D,[0.9,0.2,0.3,0.1,0.1,0.8,0.7,0.9]);
draw_graph(dag)

说明：有N=8个节点，建立有向无环图dag，并且这些点的值是离散的，这里1=False 2=True，node_sizes给出了所有状态

mk_bnet中names后的{}里面给出了各个节点的别名

利用tabular_CPD设置各个变量的边缘概率，对于A和S，定义顺序是False True；对于T、L和B这类，顺序是FF FT TF TT；对于D这类，顺序是FFF FFT FTF FTT TFF TFT TTF TTT

简单检查下A的概率



[plain] view plaincopyprint?

engine=jtree_inf_engine(bnet);

evidence=cell(1,N);

[engine,loglik]=enter_evidence(engine,evidence);

m=marginal_nodes(engine,A);

m.T()

engine=jtree_inf_engine(bnet);
evidence=cell(1,N);
[engine,loglik]=enter_evidence(engine,evidence);
m=marginal_nodes(engine,A);
m.T()

现在可以给定任意条件，然后计算概率了。

[plain] view plaincopyprint?

例如要计算任意组合条件下，个体分别得Tub、lung cancer和bronchitis的概率。下面代码计算了P（T=True|A=False,S=True,X=True,D=False）的概率

例如要计算任意组合条件下，个体分别得Tub、lung cancer和bronchitis的概率。下面代码计算了P（T=True|A=False,S=True,X=True,D=False）的概率

[plain] view plaincopyprint?

engine=jtree_inf_engine(bnet);

evidence=cell(1,N);

evidence{A}=1;

evidence{S}=2;

evidence{X}=2;

evidence{D}=1;

[engine,loglik]=enter_evidence(engine,evidence);

m=marginal_nodes(engine,T);

m.T(2)

源地址：http://blog.csdn.net/moodytong/article/details/8122327#t4