《机器学习实战Machine_Learning_in_Action》 CH10-K-均值聚类

本篇的数据和代码参见：https://github.com/stonycat/ML-in-Action

一、K-均值聚类算法

聚类是一种无监督的学习，它将相似的对象归到同一簇中，类似全自动分类。簇内的对象越相似，聚类的效果越好。K-均值聚类是每个类别簇都是采用簇中所含值的均值计算而成。聚类与分类的区别在于分类前目标已知，而聚类为无监督分类。

1.1 K-均值算法的伪代码如下：

创建k个点作为起始质心（通常随机选择）
   当任意一个点的簇分配结果发生改变时：
        对数据集中的每个点：
                对每个质心：
                计算质心与数据点之间的距离
         将数据点分配到距离其最近的簇
    对每一个簇，计算簇中所有点的均值并将均值作为质心。

基本功能函数：加载数据、计算距离、初始化k个中心三个函数。

1.2 以下为kMeans.py文件内容

from numpy import *
import numpy as np

def loadDataSet(fileName):      #general function to parse tab -delimited floats
    dataMat = []                #assume last column is target value
    fr = open(fileName)
    for line in fr.readlines():
        curLine = line.strip().split('\t')
        fltLine = list(map(float,curLine)) #map all elements to float()
        dataMat.append(fltLine)
    return dataMat

def distEclud(vecA, vecB):
    return sqrt(sum(power(vecA - vecB, 2))) #la.norm(vecA-vecB)



def randCent(dataSet, k):
    n = shape(dataSet)[1]
    centroids = mat(zeros((k,n)))#create centroid mat
    for j in range(n):#create random cluster centers, within bounds of each dimension
        #minJ = min(dataSet[:,j]) 
        #rangeJ = float(max(dataSet[:,j]) - minJ)
        minJ = float(min(dataSet[:,j]))
        rangeJ = float(max(dataSet[:,j]))- minJ
        centroids[:,j] = mat(minJ + rangeJ * random.rand(k,1))
    return centroids

def kMeans(dataSet, k, distMeas=distEclud, createCent=randCent):
    m = shape(dataSet)[0]
    clusterAssment = mat(zeros((m,2)))#create mat to assign data points 
                                      #to a centroid, also holds SE of each point
    centroids = createCent(dataSet, k)
    clusterChanged = True
    while clusterChanged:
        clusterChanged = False
        for i in range(m):#for each data point assign it to the closest centroid
            minDist = inf; minIndex = -1
            for j in range(k):
                distJI = distMeas(centroids[j,:],dataSet[i,:])
                if distJI < minDist:
                    minDist = distJI; minIndex = j
            if clusterAssment[i,0] != minIndex: clusterChanged = True
            clusterAssment[i,:] = minIndex,minDist**2
        #print (centroids)
        for cent in range(k):#recalculate centroids
            ptsInClust = dataSet[nonzero(clusterAssment[:,0].A==cent)[0]]#get all the point in this cluster
            centroids[cent,:] = mean(ptsInClust, axis=0) #assign centroid to mean 
    return centroids, clusterAssment

1.3 代码运用

# kMeans.py
import kMeans

# 加载数据
datMat = mat(kMeans.loadDataSet('testSet.txt'))
datMat[:5]
#matrix([[ 1.658985,  4.285136],
#        [-3.453687,  3.424321],
#        [ 4.838138, -1.151539],
#        [-5.379713, -3.362104],
#        [ 0.972564,  2.924086]])

#运用kMeans算法
myCentroids, clustAssing = kMeans.kMeans(datMat,4)

# 中心点
myCentroids
#matrix([[ 2.6265299 ,  3.10868015],
#        [-2.46154315,  2.78737555],
#        [-3.38237045, -2.9473363 ],
#        [ 2.80293085, -2.7315146 ]])

# 聚类信息
clustAssing[:5]
#matrix([[0.        , 2.3201915 ],
#        [1.        , 1.39004893],
#        [3.        , 6.63839104],
#        [2.        , 4.16140951],
#        [0.        , 2.7696782 ]])

# 绘制散点图
point_x = array(datMat[:,0])
point_y = array(datMat[:,1])
Centroid_x = array(myCentroids[:,0])  
Centroid_y = array(myCentroids[:,1]) 
fig, ax = plt.subplots(figsize=(10,5))
colors = array(clustAssing[:,0])
ax.scatter(point_x, point_y, s=30, c=colors, marker="o")
ax.scatter(Centroid_x,Centroid_y, s=300, c="b", marker="+", label="Centroid")
ax.legend()
ax.set_xlabel("factor1")
ax.set_ylabel("factor2")

.

二、用后处理来提高聚类性能

聚类算法中，k的值是由用户初始定义的，如何才能判断k值定义是否合适，就需要用误差来评价聚类效果的好坏，误差是各个点与其所属类别质心的距离决定的。K-均值聚类的方法效果较差的原因是会收敛到局部最小值，而且全局最小。一种评价聚类效果的方法是SSE（Sum of Squared Error）误差平方和的方法，取平方的结果是使得远离中心的点变得更加突出。
一种降低SSE的方法是增加簇的个数，即提高k值，但是违背了聚类的目标，聚类的目标是在不改变簇数目的前提下提高簇的质量。可选的改进的方法是对生成的簇进行后处理，将最大SSE值的簇划分成两个（K=2的K-均值算法），然后再进行相邻的簇合并。具体方法有两种：1、合并最近的两个质心（合并使得SSE增幅最小的两个质心）2、遍历簇 合并两个然后计算SSE的值，找到使得SSE最小的情况。
下面将使用上述技术得到更好的聚类结果方法。

三、二分K-均值算法

二分K-均值类似后处理的切分思想，初始状态所有数据点属于一个大簇，之后每次选择一个簇切分成两个簇，这个切分满足使SSE值最大程度降低，直到簇数目达到k。另一种思路是每次选择SSE值最大的一个簇进行切分。

3.1 满足使SSE值最大程度降低伪代码如下：

将所有点看成一个簇

   当簇数目小于k时
        对于每一个簇：
            计算总误差
            在给定的簇上面进行K-均值聚类（k=2）
    计算将该簇一分为二后的总误差
    选择使得误差最小的那个簇进行划分操作

函数biKmeans是上面二分K-均值聚类算法的实现，首先创建clusterAssment储存数据集中每个点的分类结果和平方误差，用centList保存所有已经划分的簇，初始状态为整个数据集。while循环不停对簇进行划分，寻找使得SSE值最大程度减小的簇并更新，添加新的簇到centList中。

3.2 以下为kMeans.py文件内容

def biKmeans(dataSet, k, distMeas=distEclud):
    m = shape(dataSet)[0]
    clusterAssment = mat(zeros((m,2)))
    centroid0 = mean(dataSet, axis=0).tolist()[0]
    centList =[centroid0] #create a list with one centroid
    for j in range(m):#calc initial Error
        clusterAssment[j,1] = distMeas(mat(centroid0), dataSet[j,:])**2
    while (len(centList) < k):
        lowestSSE = inf
        for i in range(len(centList)):
            ptsInCurrCluster = dataSet[nonzero(clusterAssment[:,0].A==i)[0],:]#get the data points currently in cluster i
            centroidMat, splitClustAss = kMeans(ptsInCurrCluster, 2, distMeas)
            sseSplit = sum(splitClustAss[:,1])#compare the SSE to the currrent minimum
            sseNotSplit = sum(clusterAssment[nonzero(clusterAssment[:,0].A!=i)[0],1])
            print ("sseSplit, and notSplit: ",sseSplit,sseNotSplit)
            if (sseSplit + sseNotSplit) < lowestSSE:
                bestCentToSplit = i
                bestNewCents = centroidMat
                bestClustAss = splitClustAss.copy()
                lowestSSE = sseSplit + sseNotSplit
        bestClustAss[nonzero(bestClustAss[:,0].A == 1)[0],0] = len(centList) #change 1 to 3,4, or whatever
        bestClustAss[nonzero(bestClustAss[:,0].A == 0)[0],0] = bestCentToSplit
        print ('the bestCentToSplit is: ',bestCentToSplit)
        print ('the len of bestClustAss is: ', len(bestClustAss))
        centList[bestCentToSplit] = bestNewCents[0,:].tolist()[0]#replace a centroid with two best centroids 
        centList.append(bestNewCents[1,:].tolist()[0])
        clusterAssment[nonzero(clusterAssment[:,0].A == bestCentToSplit)[0],:]= bestClustAss#reassign new clusters, and SSE
    return mat(centList), clusterAssment

3.3 代码运用

datMat3 = mat(kMeans.loadDataSet('testSet2.txt'))

centList,myNewAssments = kMeans.biKmeans(datMat3,3)
#sseSplit, and notSplit:  570.7227574246755 0.0
#the bestCentToSplit is:  0
#the len of bestClustAss is:  60
#sseSplit, and notSplit:  22.971771896318412 532.6598067890178
#sseSplit, and notSplit:  68.68654812621844 38.06295063565756
#the bestCentToSplit is:  1
#the len of bestClustAss is:  40

# 中心点
centList
#matrix([[-2.94737575,  3.3263781 ],
#        [-0.45965615, -2.7782156 ],
#        [ 2.93386365,  3.12782785]])

# 聚类信息
myNewAssments[:5]
#matrix([[2.        , 0.14546105],
#        [0.        , 0.68021383],
#        [1.        , 1.02184582],
#        [2.        , 1.3454876 ],
#        [0.        , 1.35376464]])

# 绘制散点图
point_x = array(datMat3[:,0])
point_y = array(datMat3[:,1])
Centroid_x = array(centList[:,0])  
Centroid_y = array(centList[:,1]) 
fig, ax = plt.subplots(figsize=(10,5))
colors = array(myNewAssments[:,0])
ax.scatter(point_x, point_y, s=30, c=colors, marker="o")
ax.scatter(Centroid_x,Centroid_y, s=300, c="b", marker="+", label="Centroid")
ax.legend()
ax.set_xlabel("factor1")
ax.set_ylabel("factor2")