CEP.CEPListener.java Source code

Introduction

Here is the source code for CEP.CEPListener.java

Source

/*
 * To change this license header, choose License Headers in Project Properties.
 * To change this template file, choose Tools | Templates
 * and open the template in the editor.
 */
package CEP;

import com.espertech.esper.client.EventBean;
import com.espertech.esper.client.UpdateListener;
import java.util.logging.Level;
import java.util.logging.Logger;
import weka.classifiers.Evaluation;
import weka.classifiers.meta.FilteredClassifier;
import weka.core.Instance;
import weka.core.Instances;
import weka.filters.unsupervised.attribute.Remove;

/**
 *
 * @author gs023850
 */
public class CEPListener implements UpdateListener {

    Instances data, train;
    boolean training;
    FilteredClassifier tree;
    int cumulative, sampleSize, windowSize;
    double accuracy;

    public CEPListener(int windowSize) throws InterruptedException {
        tree = CreateClassifier();
        this.windowSize = windowSize;
        data = null;
        cumulative = 0;
        training = true;
        sampleSize = 200;
    }

    public void SetLabels(HeaderManager labels) {
        // this.labels = labels;
    }

    public void update(EventBean[] newData, EventBean[] oldData) {

        System.out.println("Event received: " + newData[0].getUnderlying());

        if (newData.length > 0) {
            try {
                if (training) {
                    if (train == null) {
                        train = HeaderManager.GetEmptyStructure();
                    }
                    for (EventBean bean : newData) {
                        Object inst = bean.getUnderlying();
                        train.add((Instance) inst);
                    }
                    if (train.size() >= sampleSize) {
                        tree.buildClassifier(train);
                        training = false;
                    }
                } else {
                    if (data == null) {
                        data = HeaderManager.GetStructure();
                    }

                    data = SetDuration(data);
                    cumulative += data.size();

                    for (EventBean bean : newData) {
                        Object inst = bean.getUnderlying();
                        data.add((Instance) inst);
                    }
                    for (int i = data.numInstances() - newData.length; i < data.numInstances(); i++) {
                        double pred = tree.classifyInstance(data.instance(i));
                        System.out.print("ID: " + data.instance(i).value(0));
                        System.out.print(
                                ", actual: " + data.classAttribute().value((int) data.instance(i).classValue()));
                        System.out.println(", predicted: " + data.classAttribute().value((int) pred));
                        Evaluation eval = new Evaluation(data);
                        if ((accuracy = eval.rootMeanSquaredError()) < 0.7) {

                            training = true;
                            train.clear();
                            train = null;
                        }
                        System.out.print("Accuracy: " + accuracy);
                    }
                }
            } catch (InterruptedException ex) {
                Logger.getLogger(CEPListener.class.getName()).log(Level.SEVERE, null, ex);
            } catch (Exception ex) {
                Logger.getLogger(CEPListener.class.getName()).log(Level.SEVERE, null, ex);
            }

        }
    }

    Instances SetDuration(Instances in) throws InterruptedException {
        Instances out = HeaderManager.GetEmptyStructure();
        for (Instance inst : in) {

            double time = inst.toDoubleArray()[5];
            if (time > 20) {
                inst.setValue(5, time - windowSize);
                out.add(inst);
            }
        }
        return out;

    }

    FilteredClassifier CreateClassifier() {
        Remove rm = new Remove();
        //rm.setAttributeIndices("1");  // remove 1st attribute
        PART c = new PART();
        // meta-classifier
        FilteredClassifier fc = new FilteredClassifier();
        fc.setFilter(rm);
        fc.setClassifier(c);
        return fc;
    }
}