K-Means聚类分析算法C语言实现

#include<stdio.h> #include<stdlib.h> #include<math.h> #define K 2 #define MAX 100000 typedef struct{ double featureVector[10]; int classification; }sampleNode; int m, n; sampleNode *sn; sampleNode z[K+1]; sampleNode zz[K+1]; int classAmount[K+1]; //classAmount[k]表示当前类K的样本数量 int center[K+1]; void inputSample(){ int i, j; printf("请输入样本的数量和样本的特征向量的属性个数（用空格隔开）：\n"); scanf("%d%d", &m, &n); sn = (sampleNode*)malloc(sizeof(sampleNode)*(m+1)); for(i=0; i<m; i++){ sn[i].classification = 0; } printf("\n请依次输入每个样本的特征值，特征值用空格隔开，样本用回车符隔开:\n"); for(i=1; i<=m; i++){ printf("第%2d个样本：", i); for(j=1; j<=n; j++){ scanf("%lf", &sn[i].featureVector[j]); } } } double getDistance(sampleNode a, sampleNode b){ int i; double dis; dis = 0.0; for(i=1; i<=n; i++){ dis += pow(a.featureVector[i]-b.featureVector[i],2); } return sqrt(dis); } void classify(){ double min; double temp[K+1]; double distance; int i, j; int c; c = 1; for(i=1; i<=m; i++){ for(j=1; j<=K; j++){ temp[j] = getDistance(sn[i], z[j]); } min = MAX; for(j=1; j<=K; j++){ if(temp[j] < min){ min = temp[j]; c = j; } } sn[i].classification = c; } } void getCurrentAmount(){ int i; for(i=0; i<=K; i++){ classAmount[i] = 0; } for(i=1; i<=m; i++){ classAmount[sn[i].classification]++; } } void reCalCenter(){ int i; int j; double value[K+1]; int k; getCurrentAmount(); for(i=1; i<=n; i++){ for(k=0; k<=K; k++){ value[k] = 0.0; } for(j=1; j<=m; j++){ value[sn[j].classification] += sn[j].featureVector[i]; } for(k=1; k<=K; k++){ zz[k].featureVector[i] = value[k]/classAmount[k]; } } } int judgeCenterEquality(){ int i, j; int flag; flag = 1; for(i=1; i<=n; i++){ for(j=1; j<=K; j++){ if(z[j].featureVector[i] != zz[j].featureVector[i]){ flag = 0; break; } } if(!flag){ break; } } return flag; } void centerExchange(){ int i, j; for(j=1; j<=K; j++){ for(i=1; i<=n; i++){ z[j].featureVector[i] = zz[j].featureVector[i]; z[j].featureVector[i] = zz[j].featureVector[i]; } } } void printResult(){ int i, j; for(i=1; i<=K; i++){ printf("属于类别%d的样本如下：\n", i); for(j=1; j<=m; j++){ if(sn[j].classification == i){ printf("X%d ", j); } } printf("\n"); } printf("\n\n"); } int main(){ int i; int center[K+1]; inputSample(); printf("\n请输入%d个初始聚类中心，用空格隔开：\n", K); for(i=1; i<=K; i++){ scanf("%d", &center[i]); } for(i=1; i<=K; i++){ z[i] = sn[center[i]]; } i = 1; classify(); reCalCenter(); while(!judgeCenterEquality()){ centerExchange(); classify(); reCalCenter(); i++; } printResult(); printf("\n迭代次数为：%d\n", i); return 0; }