k_means算法C++实现，改为面向对象

画的类图如下：

aaarticlea/png;base64,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" alt="" width="782" height="392" />

分为两个类kMeans和pointClass类，kMeans两个成员变量：节点总个数和最终聚类个数。pointClass类包含结构体point。

具体代码如下：

kMeans.h

class kMeans{
protected:
    int numOfPoint, numOfCenter;
public:
    kMeans();
    void k_means();
};

kMeans.cpp

#include "kMeans.h"
#include "point.h"
#include <iostream>
using namespace std;

kMeans::kMeans()
{
    cout << "分别输入数据点和最终聚类的个数: ";
    cin >> numOfPoint >> numOfCenter;
}

void kMeans::k_means()
{
    pointClass pc(numOfPoint, numOfCenter);
    pc.InitCenter(numOfPoint, numOfCenter);
    bool b = true;
    while (b){
        pc.setPoint(numOfPoint, numOfCenter);
        if (pc.getError()){
            break;
        }
        pc.getNewCenter(numOfPoint, numOfCenter);
        b = pc.IsEnd(numOfCenter);
        pc.resetCenterOld(numOfCenter);
    }
    if (b) {
        cout << "聚类操作无法完成！！" << endl;
    }else{
        pc.ExportData(numOfPoint, numOfCenter);
    }
}

point.h

struct point{
    double x1, x2, x3, x4;
    int flag;
};

class pointClass{
protected:
    double errorOld, errorNew;
    point *pList, *centerListNew, *centerListOld;
    double GetDistance(point point1, point point2);
    bool GetExist(int xm, int centerList[], int n);
public:
    pointClass (int numOfPoint, int numOfCenter);
    void InitCenter (int numOfPoint, int numOfCenter);
    void setPoint (int numOfPoint, int numOfCenter);
    bool getError();
    void getNewCenter (int numOfPoint, int numOfCenter);
    bool IsEnd (int numOfCenter);
    void ExportData (int numOfPoint, int numOfCenter);
    void resetCenterOld (int numOfCenter);
    ~pointClass ();
};

point.cpp

#include <iostream>
#include <fstream>
#include <cmath>
#include <ctime>
#include "point.h"
using namespace std;

double pointClass::GetDistance(point point1, point point2)
{
    return pow(pow(point1.x1 - point2.x1, ) + pow(point1.x2 - point2.x2, ) + pow(point1.x3 - point2.x3, ) + pow(point1.x4 - point2.x4, ), 0.5);
}

bool pointClass::GetExist(int xm, int CentIndex[], int n)
{
    bool b = false;
    for (int i = ; i < n; i++){
        if (xm == CentIndex[i]){
            b = true;
            break;
        }
    }
    return b;
}

pointClass::pointClass(int numOfPoint, int numOfCenter)
{
    pList = new point[numOfPoint];
    centerListOld = new point[numOfCenter];
    centerListNew = new point[numOfCenter];
    ifstream ifile("D:\\IrisData.txt");
    if (!ifile.is_open()){
        cerr << "file" << endl;
        exit();
    }
    int i = ;
    while (i < numOfPoint){
        ifile >> pList[i].x1 >> pList[i].x2 >> pList[i].x3 >> pList[i].x4;
        pList[i].flag = ;
        i++;
    }
    ifile.close();

    errorNew = , errorOld = ;
    for (i = ; i < numOfCenter; i++){
        centerListOld[i].x1 = ;
        centerListOld[i].x2 = ;
        centerListOld[i].x3 = ;
        centerListOld[i].x4 = ;
        centerListNew[i].x1 = ;
        centerListNew[i].x2 = ;
        centerListNew[i].x3 = ;
        centerListNew[i].x4 = ;
        centerListNew[i].flag = ;
        centerListOld[i].flag = ;
    }
}

void pointClass::InitCenter(int numOfPoint, int numOfCenter)
{
    int xm, i;
    int *CenterIndex = new int[numOfCenter];
    srand((unsigned)time());
    for (i = ; i < numOfCenter; i++){
        do {
            xm = rand() % numOfPoint;
        } while (GetExist(xm, CenterIndex, i));
        CenterIndex[i] = xm;
    }

    for (i = ; i < numOfCenter; i++){
        centerListOld[i] = pList[CenterIndex[i]];
    }
}

void pointClass::setPoint(int numOfPoint, int numOfCenter)
{
    errorNew = ;
    for (int i = ; i < numOfPoint; i++){
        int flagi = ;
        double distance = GetDistance(pList[i], centerListOld[]);
        for (int j = ; j < numOfCenter; j++){
            double tmp = GetDistance(pList[i], centerListOld[j]);
            if (tmp < distance){
                tmp = distance;
                flagi = j;
            }
        }
        pList[i].flag = flagi;
        errorNew = GetDistance(pList[i], centerListOld[flagi]);
    }
}

bool pointClass::getError()
{
    bool b = false;
    if (errorOld !=  && errorNew >= errorOld){
        b = true;
    }
    return b;
}

void pointClass::getNewCenter(int numOfPoint, int numOfCenter)
{
    for (int i = ; i < numOfCenter; i++){
        centerListNew[i].x1 = ;
        centerListNew[i].x2 = ;
        centerListNew[i].x3 = ;
        centerListNew[i].x4 = ;
        centerListNew[i].flag = ;
    }
    for (int i = ; i < numOfPoint; i++){
        centerListNew[pList[i].flag].x1 += pList[i].x1;
        centerListNew[pList[i].flag].x2 += pList[i].x2;
        centerListNew[pList[i].flag].x3 += pList[i].x3;
        centerListNew[pList[i].flag].x4 += pList[i].x4;
        centerListNew[pList[i].flag].flag++;
    }
    for (int i = ; i < numOfCenter; i++){
        centerListNew[i].x1 = centerListNew[i].x1 / centerListNew[i].flag;
        centerListNew[i].x2 = centerListNew[i].x2 / centerListNew[i].flag;
        centerListNew[i].x3 = centerListNew[i].x3 / centerListNew[i].flag;
        centerListNew[i].x4 = centerListNew[i].x4 / centerListNew[i].flag;
        centerListNew[i].flag = ;
    }
}

void pointClass::resetCenterOld(int numOfCenter)
{
    for (int i = ; i < numOfCenter; i++){
        centerListOld[i] = centerListNew[i];
    }
}

bool pointClass::IsEnd(int numOfCenter)
{
    bool b = false;
    for (int i = ; i < numOfCenter; i++){
        if (GetDistance(centerListNew[i],centerListOld[i]) > ){
            b = true;
            break;
        }
    }
    return b;
}

void pointClass::ExportData(int numOfPoint, int numOfCenter)
{
    ofstream ofile("D:\\kMeansResult.txt");
    cout << "本次误差是：" << errorNew << endl;
    ofile << "本次误差是：" << errorNew << endl;
    for (int j = ; j < numOfCenter; j++){
        ofile << "第" << j+ <<"类：" << endl;
        for (int i = ; i < numOfPoint; i++){
            if (pList[i].flag == j){
                ofile << pList[i].x1 << " " << pList[i].x2 << " " << pList[i].x3 << " " <<pList[i].x4 << endl;
            }
        }
    }
}

pointClass::~pointClass()
{
    delete[] pList;
    delete[] centerListOld;
    delete[] centerListNew;
}

主函数main.cpp

#include <iostream>
#include "kMeans.h"
using namespace std;

int main()
{
    kMeans km;
    km.k_means();
    return ;
}