kmeans算法及python代码参考

算法介绍和分析请参考：http://hi.baidu.com/chenjinandy/blog/item/5c4d5344b607c3248794739e.html

# -*- coding:gb2312 -*-

'''

Created on 2011-10-21

 @author: chenjinandy

'''

from math import sqrt

def pearson(v1,v2):

  # Simple sums

  sum1=sum(v1)

  sum2=sum(v2)

  # Sums of the squares

  sum1Sq=sum([pow(v,2) for v in v1])

  sum2Sq=sum([pow(v,2) for v in v2])    

  # Sum of the products

  pSum=sum([v1[i]*v2[i] for i in range(len(v1))])

  # Calculate r (Pearson score)

  num=pSum-(sum1*sum2/len(v1))

  den=sqrt((sum1Sq-pow(sum1,2)/len(v1))*(sum2Sq-pow(sum2,2)/len(v1)))

  if den==0: return 0

  return 1.0-num/den  

v1=[1.22,1.33,4.55,6.45,6.31,4.12]

v2=[1.22,1.33,4.55,6.45,6.31,4.12]

print pearson(v1,v2) 

dataset=[(1,0),(0,1),(1,1),(2,1),(1,2),(2,2),(3,2),(6,6),(7,6),(8,6),(6,7),(7,7),(8,7),(9,7),(7,8),(8,8),(9,8),(8,9),(9,9)] 

print dataset

#kmeans 的pythondex实现 

def isequal(v1,v2):

    if (v1[0]*10==v2[0]*10)and(v1[1]*10==v2[1]*10):

        return 1

    else:

        return 0

def order(dataset,k):

    type=[]

    temp=0

    for i in range(len(dataset)):

        for j in range(k):

            if pearson(dataset[i],dataset[j])<pearson(dataset[i],dataset[temp]):

                                               temp=j

        type.append(temp)

        print dataset[i]

        print "被归为第 %s" %temp+"类"    

    return type

def newcenter(dataset,num,type):

    cnt=0

    sumx=0.0

    sumy=0.0

    point=(0.0,0.0)

    for i in range(len(dataset)):

        if type[i]==num:

            sumx+=dataset[i][0]

            sumy+=dataset[i][1]

            cnt+=1

    sumx=sumx/cnt

    sumy=sumy/cnt

    point=(sumx,sumy)

    return point

def cluster(dataset,k):

    center=[]

    oldcenter=[]

    for i in range(k):

       # print i,dataset[i]

        cen=dataset[i]

        center.append(cen)

        oldcen=(0.0,0.0)

        oldcenter.append(oldcen)

    cnt=0

    while cnt!=k:

        type=order(dataset,k)

        for i in range(k):

            center[i]=newcenter(dataset,i,type)

            if isequal(center[i],oldcenter[i]):

                cnt+=1

            else:

                oldcenter[i]=center[i]               

    for i in range(k):

        print "属于类别 %s" %i+"的点有如下："

        for j in range(len(dataset)):

            if type[j]==i:

                print dataset[j]

for i in range(len(dataset)):

    print dataset[i]

cluster(dataset,3)