mylib.cc

一个增量文本聚类的算法。 参考文献： Wai-chiu Wong, Ada Wai-chee Fu, Incremental Document Clustering for Web Page Cl

#include <stdio.h>
#include <stdlib.h>
#include <ctype.h>
#include <string.h>
#include <iostream.h>
#include <time.h>
#include "DC.h"
#include "MyLib.h"


void dummy()
{
    printf("I'm a dummy.\n");
}

int ranGen(int seed,int num)
{
    srand((time(NULL))+(seed*121)*37);
    return (rand()%num);
}

double man_sim(DC *dc1,DC *dc2)
{
    double s,dist,w1,w2;
    int nf;
    nf = dc1->numFeat;
    dist = 0.0;
    for(int i=0;i<nf;i++){
        w1 = ((double) dc1->W[i])/((double) dc1->N);
        w2 = ((double) dc2->W[i])/((double) dc2->N);
        dist = dist + fabs(w1-w2);
    }
    s = 1.0-(dist/((double) nf));
    return s;
}

double cos_sim(DC *dc1,DC *dc2)
{
    double s,w1,w2,prod,norm1,norm2;
    int nf;
    nf = dc1->numFeat;
    prod = 0.0;
    norm1 = 0.0;
    norm2 = 0.0;
    for(int i=0;i<nf;i++){
        w1 = ((double) dc1->W[i])/((double) dc1->N);
        w2 = ((double) dc2->W[i])/((double) dc2->N);
        prod = prod+(w1*w2);
        norm1 = norm1+(w1*w1);
        norm2 = norm2+(w2*w2);
    }
    s = prod/sqrt(norm1*norm2);
    return s;
}

double euc_sim(DC *dc1,DC *dc2)
{
    double s,dist,w1,w2,diff;
    int nf;
    nf = dc1->numFeat;
    dist = 0.0;
    for(int i=0;i<nf;i++){
        w1 = ((double) dc1->W[i])/((double) dc1->N);
        w2 = ((double) dc2->W[i])/((double) dc2->N);
        diff = w1-w2;
        dist = dist+(diff*diff);
    }
    s = 1.0-(sqrt(dist)/((double) nf));
    return s;
}

double sim(int simID,DC *dc1,DC *dc2)
{
    if(simID == 1){
	return cos_sim(dc1,dc2);
    }else if(simID == 2){
	return euc_sim(dc1,dc2);
    }else{
	return man_sim(dc1,dc2);
    }
}