集成学习之 AdaBoosting-1带权数据集

AdaBoosting 基本思想

首先了解一下AdaBoosting 。这是一个把多个弱分类器（分类精度略高于50%）叠加成为强分类器（分类精度高于90%）的算法。

其步骤如下：

1. 初始化训练数据的权值。
2. 训练弱分类器。如某个训练样本能被弱分类器准确分类，那么在构造下一个训练集时，其对应的权值减小。若被错误分类，那么权重增大。权值更新过的训练集会被用于训练下一个分类器。
3. 组合所有的弱分类器为一个强分类器。加大分类误差率小的弱分类器的权重，使其在最终的分类函数中起着较大的决定作用，而降低分类误差率大的弱分类器的权重，使其在最终的分类函数中起着较小的决定作用。

关于权值调整

初始权值：若共有N个训练样本，则每个训练样本的权值为 ，权值和为1.0。

调整权值步骤：

1. 若第i个样本分类正确，权值调整为：。若分类错误则调整为：。其中为参数，为样本原本的权值。
2. 为保证最终权值总和为1.0，第i个样本的权值应为：

输入：arff数据集

输出：初始权值；调整后的权值；保证最终权值总和为1.0的权值。

优化目标：可能没有优化目标。

代码如下：

package knn5;
import java.io.FileReader;
import java.util.Arrays;
import weka.core.*;

public class WeightedInstances extends Instances{

	private static final long serialVersionUID = 11087456L;
	private double[] weights;
	
	public WeightedInstances(FileReader paraFileReader) throws Exception{
		super(paraFileReader);
		setClassIndex(numAttributes() - 1);
		
		weights = new double[numInstances()];
		double tempAverage = 1.0/numInstances();
		for(int i = 0; i < weights.length; i++) {
			weights[i] = tempAverage;
		} // Of for i
		System.out.println("Instances weights are: " + Arrays.toString(weights));
	} // Of the first constructor

	public WeightedInstances(Instances paraInstances) {
		super(paraInstances);
		setClassIndex(numAttributes() - 1);
		
		weights = new double[numInstances()];
		double tempAverage = 1/numInstances();
		for(int i = 0; i < weights.length; i++) {
			weights[i] = tempAverage;
		} // Of for i
		System.out.println("Instances weights are: " + Arrays.toString(weights));
	} // Of the second constructor

	public double getWeight(int paraIndex) {
		return weights[paraIndex];
	}//Of getWeight
	
	public void adjustWeights(boolean[] paraCorrectArray, double paraAlpha) {
		double tempIncrease = Math.exp(paraAlpha);
		
		double tempWeightsSum = 0;
		for(int i =0; i < weights.length; i++) {
			if (paraCorrectArray[i]) {
				weights[i] /= tempIncrease;
			} else {
				weights[i] *= tempIncrease;
			} // Of if
			tempWeightsSum += weights[i];
		} // Of for i
		for (int i = 0; i < weights.length; i++) {
			weights[i] /= tempWeightsSum;
		} // Of for i
		System.out.println("After adjusting, instances weights are: " + Arrays.toString(weights));
	} // Of adjustWeights
	
	public void adjustWeightsTest() {
		boolean[] tempCorrectArray = new boolean[numInstances()];
		for (int i = 0; i < tempCorrectArray.length / 2; i++) {
			tempCorrectArray[i] = true;
		} // Of for i

		double tempWeightedError = 0.3;

		adjustWeights(tempCorrectArray, tempWeightedError);

		System.out.println("After adjusting");

		System.out.println(toString());
	} // Of adjustWeightsTest
	public String toString() {
		String resultString = "I am a weighted Instances object.rn" + "I have " + numInstances() + " instances and "
				+ (numAttributes() - 1) + " conditional attributes.rn" + "My weights are: " + Arrays.toString(weights)
				+ "rn";

		return resultString;
	} // Of toString

	public static void main(String[] args) {
		WeightedInstances tempWeightedInstances = null;
		String tempFilename = "C:\\Users\\ASUS\\Desktop\\文件\\iris.arff";
		try {
			FileReader tempFileReader = new FileReader(tempFilename);
			tempWeightedInstances = new WeightedInstances(tempFileReader);
			tempFileReader.close();
		} catch (Exception exception1) {
			System.out.println("Cannot read the file: " + tempFilename + "rn" + exception1);
			System.exit(0);
		} // Of try

		System.out.println(tempWeightedInstances.toString());

		tempWeightedInstances.adjustWeightsTest();
	} // Of main

} // Of class WeightedInstances

运行截图：