分类问题 Python 朴素贝叶斯 屏蔽系统（学习笔记）

from numpy import *
import re
import feedparser

'''包含所有文档不带重复词的list'''
def createVocabList(dataSet):
    vocabSet=set([])#不带重复值的list
    for document in dataSet:
        vocabSet=vocabSet|set(document)#不带重复词的并集
    return list(vocabSet)

'''判断某个词条在文档中是否出现'''
def setOfWords(vocabList,inputSet):
    returnVec=[0]*len(vocabList)
    for word in inputSet:
        if word in vocabList:
            returnVec[vocabList.index(word)]=1
        else:
            print("单词：%s不在我的词汇里面！"%word)
    return returnVec #返回0/1的列表

'''朴素贝叶斯分类训练函数'''
def trainNB(trainMatrix,trainCategory):
    numTraindocs=len(trainMatrix)
    numWords=len(trainMatrix[0])
    pAbusive=sum(trainCategory)/float(numTraindocs)#1类的概率
    p0Num=ones(numWords)
    p1Num=ones(numWords)
    p0Denom=2.0
    p1Denom=2.0
    for i in range(numTraindocs):
        if trainCategory[i]==1:
            p1Num+=trainMatrix[i]#积累单个词出现的次数
            p1Denom+=sum(trainMatrix[i])#积累总单词出现的次数
        else:
            p0Num+=trainMatrix[i]
            p0Denom+=sum(trainMatrix[i])
            
    p0Vect=log(p0Num/p0Denom)#0类每个单词出现的频率
    p1Vect=log(p1Num/p1Denom)#1类每个单词出现的频率
    return p0Vect,p1Vect,pAbusive

'''定义分类'''
def classifyNB(vec2Classify,p0Vec,p1Vec,pClass1):
    p1=sum(vec2Classify*p1Vec)+log(pClass1)
    p0=sum(vec2Classify*p0Vec)+log(1.0-pClass1)
    if p1>p0:
        return 1
    else:
        return 0

#数据
def loadDataSet():
    postingList=[['my', 'dog', 'has', 'flea', 'problems', 'help', 'please'],
                 ['maybe', 'not', 'take', 'him', 'to', 'dog', 'park', 'stupid'],
                 ['my', 'dalmation', 'is', 'so', 'cute', 'I', 'love', 'him'],
                 ['stop', 'posting', 'stupid', 'worthless', 'garbage'],
                 ['mr', 'licks', 'ate', 'my', 'steak', 'how', 'to', 'stop', 'him'],
                 ['quit', 'buying', 'worthless', 'dog', 'food', 'stupid']]
    classVec = [0,1,0,1,0,1]    #1表示侮辱类，0表示不属于
    return postingList,classVec #词条切分后的分档和类别标签
    
#预测
def testNB():
    resultList=["非侮辱类","侮辱类"]
    postingList,classVec=loadDataSet()#导入数据
    vocabSet=createVocabList(postingList)#不重复列表
    
    trainingSet=len(classVec)
    trainMat=[]
    for docIndex in range(trainingSet):
        trainMat.append(setOfWords(vocabSet,postingList[docIndex]))
    p0,p1,pA=trainNB(array(trainMat),array(classVec))
    
    testEntry=['love','my','dalmation']
    print("将'love','my','dalmation'分为：",resultList[classifyNB(setOfWords(vocabSet,testEntry),p0,p1,pA)])
    testEntry=['stupid','garbage']
    print("将'stupid','garbage'分为：",resultList[classifyNB(setOfWords(vocabSet,testEntry),p0,p1,pA)])

testNB()

输出结果：

将’love’,‘my’,‘dalmation’分为： 非侮辱类
将’stupid’,'garbage’分为： 侮辱类