java通过wekaapi构建bayes网

Question

我想用一个现有的数据集构建一个贝叶斯网络（或贝叶斯网络），有没有人用WEKAAPI学习贝叶斯网络的演示代码

Answer 1

//==============================学习贝叶斯网络（结构和CPT）===================== //K2学习者=新K2（）；//集合搜索算法1

    //MultiNomialBMAEstimator estimator = new MultiNomialBMAEstimator(); //set estimator 1
    //estimator.setUseK2Prior(true);

    //BMAEstimator estimator = new BMAEstimator(); //set estimator 2

    SimpleEstimator estimator = new SimpleEstimator(); //set estimator 3
    SimulatedAnnealing learner = new SimulatedAnnealing(); // set search algorithm 2    

    Evaluation evaluation = new Evaluation(ins);

    EditableBayesNet bnBayesNet = new EditableBayesNet(ins);
    bnBayesNet.initStructure();

    learner.buildStructure(bnBayesNet, ins);
    estimator.estimateCPTs(bnBayesNet);

    //=================== visualize the bayes net ====================
    GraphVisualizer graphVisualizer = new GraphVisualizer();
    graphVisualizer.readBIF(bnBayesNet.graph());

    final JFrame jFrame = new JFrame("Weka Classifier Graph Visualizer: Bayes net");
    jFrame.setSize(500, 400);
    jFrame.getContentPane().setLayout(new BorderLayout());
    jFrame.getContentPane().add(graphVisualizer, BorderLayout.CENTER);
    jFrame.addWindowFocusListener(new WindowAdapter() {
        public void windowClosing(WindowEvent e) {
            jFrame.dispose();               
        }
    });
    jFrame.setVisible(true);
    graphVisualizer.layoutGraph();