package PCI.ASQ.classification;
import java.util.ArrayList;
import java.util.Iterator;
import java.util.List;

import PCI.ASQ.basic.RatioMatrix;
//这个函数用于模糊概率集平均值采样分类
//思想：贝叶斯 模糊数学 统计于概率论
//实现：罗瑶光
//作者,著作权人： 罗瑶光，浏阳
public class FuzzProbabailityClassification{
  public static RatioMatrix getSimilarFuzzSet(RatioMatrix input, List<RatioMatrix> groups) {
    double shortestSumRight=0;
    boolean isFirst= true;
    int key = 0;
    int i= 0;
    //成功集
    Iterator<RatioMatrix> iterators= groups.iterator();
    while(iterators.hasNext()) {
      RatioMatrix ratio= iterators.next();
      double rightRightMean= ratio.getRightRightRatio();
      double rightErrorMean= ratio.getRightErrorRatio();
      double errorRightMean= ratio.getErrorRightRatio();
      double errorErrorMean= ratio.getErrorErrorRatio();
      double predictionRightRight= Math.abs(input.getRightRightRatio()- rightRightMean);
      double predictionRightError= Math.abs(input.getRightErrorRatio()- rightErrorMean);
      double predictionErrorRight= Math.abs(input.getErrorRightRatio()- errorRightMean);
      double predictionErrorError= Math.abs(input.getErrorErrorRatio()- errorErrorMean);
      double tempSumRight= predictionRightRight+ predictionRightError+ predictionErrorRight
          + predictionErrorError;
      if(true== isFirst) {
        isFirst= false;
        shortestSumRight= tempSumRight;
        key=i;
      }else {
        if(shortestSumRight> tempSumRight) {
          shortestSumRight= tempSumRight;
          key= i;
        }
      }
      i++;
    }
    return groups.get(key);  
  }  
  
  public static List<RatioMatrix> getSimilarFuzzSetListWithScale(RatioMatrix input, List<RatioMatrix> groups
      , double scale) {
    List<RatioMatrix> output= new ArrayList<>();
    Iterator<RatioMatrix> iterators= groups.iterator();
    while(iterators.hasNext()) {
      RatioMatrix ratio= iterators.next();
      double rightRightMean= ratio.getRightRightRatio();
      double rightErrorMean= ratio.getRightErrorRatio();
      double errorRightMean= ratio.getErrorRightRatio();
      double errorErrorMean= ratio.getErrorErrorRatio();
      double predictionRightRight= Math.abs(input.getRightRightRatio()- rightRightMean);
      double predictionRightError= Math.abs(input.getRightErrorRatio()- rightErrorMean);
      double predictionErrorRight= Math.abs(input.getErrorRightRatio()- errorRightMean);
      double predictionErrorError= Math.abs(input.getErrorErrorRatio()- errorErrorMean);
      double tempSumRight= predictionRightRight+ predictionRightError+ predictionErrorRight
          + predictionErrorError;
      if(tempSumRight< scale) {
        output.add(ratio);
      }
    }
    return output;  
  }  
}