三、（3）PCA降3维下的K-means聚类可视化

三、（3）PCA降3维下的K-means聚类可视化

完整代码如下：

# -*- coding: utf-8 -*-
"""
Created on Wed May 15 11:40:27 2019

@author: sun
"""


import codecs
import matplotlib.pyplot as plt
from sklearn.feature_extraction.text import TfidfTransformer  
from sklearn.feature_extraction.text import CountVectorizer


if __name__ == "__main__":
    
    #文档预料 空格连接
    corpus = []
    
    #读取预料 一行预料为一个文档
    for line in open('聚类4类.txt', 'r',encoding='UTF-8').readlines():
        #print line
        corpus.append(line.strip())
    

    #将文本中的词语转换为词频矩阵 矩阵元素a[i][j] 表示j词在i类文本下的词频
    vectorizer = CountVectorizer(min_df=10)
 
    #该类会统计每个词语的tf-idf权值
    transformer = TfidfTransformer()
 
    #第一个fit_transform是计算tf-idf 第二个fit_transform是将文本转为词频矩阵
    tfidf = transformer.fit_transform(vectorizer.fit_transform(corpus))

    #tfidf = vectorizer.fit_transform(corpus).toarray() #LI
    
    #获取词袋模型中的所有词语  
    word = vectorizer.get_feature_names()
    
    #将tf-idf矩阵抽取出来，元素w[i][j]表示j词在i类文本中的tf-idf权重
    weight = tfidf.toarray()
 
    #打印特征向量文本内容
    #print( 'Features length: ' + str(len(word)))
    resName = "BHTfidf_Result.txt"
    result = codecs.open(resName, 'w', 'utf-8')
    for j in range(len(word)):
        result.write(word[j] + ' ')
    result.write('\r\n\r\n')
 
    #打印每类文本的tf-idf词语权重，第一个for遍历所有文本，第二个for便利某一类文本下的词语权重  
    for i in range(len(weight)):
        #print u"-------这里输出第", i, u"类文本的词语tf-idf权重------"  
        for j in range(len(word)):
            #print weight[i][j],
            result.write(str(weight[i][j]) + ' ')
        result.write('\r\n\r\n')
 
    result.close()
 
    print( 'Start Kmeans:')
    from sklearn.cluster import KMeans
    clf = KMeans(n_clusters=4)   #科技 医学 汽车 国家
    s = clf.fit(weight)

    #每个样本所属的簇
    label = []              
    #print(clf.labels_)
    i = 1
    while i <= len(clf.labels_):
        #print( clf.labels_[i-1])
        label.append(clf.labels_[i-1])
        i = i + 1
 
    from sklearn.decomposition import PCA
    pca = PCA(n_components=3)             #输出三维
    newData = pca.fit_transform(weight)   #载入N维

    x = [n[0] for n in newData]
    y = [n[1] for n in newData]
    z = [n[2] for n in newData]
    x1, y1,z1 = [], [],[]
    x2, y2,z2 = [], [],[] 
    x3, y3,z3 = [], [],[]
    x4, y4,z4 = [], [],[]


    #分布获取类标为0、1、2、3的数据 赋值给(x1,y1) (x2,y2) (x3,y3) (x4,y4)
    i = 0  
    while i < len(newData):  
        if y_pred[i]==0:  
            x1.append(newData[i][0])  
            y1.append(newData[i][1])
            z1.append(newData[i][2])
        elif y_pred[i]==1:  
            x2.append(newData[i][0])  
            y2.append(newData[i][1]) 
            z2.append(newData[i][2])
        elif y_pred[i]==2:  
            x3.append(newData[i][0])  
            y3.append(newData[i][1])
            z3.append(newData[i][2])
        elif y_pred[i]==3:  
            x4.append(newData[i][0])  
            y4.append(newData[i][1])
            z4.append(newData[i][2])
        i = i + 1
 
    fig=plt.figure()
    ax=fig.add_subplot(111,projection='3d')
    ax.scatter(x1,y1,z1,c='m')
    ax.scatter(x2,y2,z2,c='r')
    ax.scatter(x3,y3,z3,c='g')
    ax.scatter(x4,y4,z4,c='c')
    #颜色包括 g b y r c m k
    p#lt.show()

可视化图形如下：