【机器学习】朴素贝叶斯-02

心得体会

　　1交叉验证：从训练的数据里随机抽取作为测试集

# 4-6朴素贝叶斯过滤垃圾邮件
#朴素贝叶斯交叉验证
def textParse(bigString):
    import re
    listOfTokens=re.split('\W+',bigString)
    return [tok.lower() for tok in listOfTokens if len(tok)>2]

def spamTest():
    docList=[]
    classList=[]
    fullText=[]
    #存入数据
    for i in range(1,26):
        wordList=textParse(open('E:/Python/《机器学习实战》代码/Ch04/spam/%d.txt'%i).read())
        docList.append(wordList)
        fullText.append(wordList)
        classList.append(1)
        wordList=textParse(open('E:/Python/《机器学习实战》代码/Ch04/ham/%d.txt'%i).read())
        docList.append(wordList)
        fullText.append(wordList)
        classList.append(0)
    # 随机建立测训练集
    vocabList=createVocabList(docList)  #生成单词向量
    trainingSet=list(range(50))
    testSet=[]
    for i in range(10):
        randIndex=int(random.uniform(0,len(trainingSet)))#生成trainingSet范围内的随机数
        testSet.append(trainingSet[randIndex])#将该训练数据加入训练集
        del(trainingSet[randIndex])#将training中该数据删除，以免重复使用同一数据
    trainMat=[]
    trainClasses=[]
    for docIndex in trainingSet:
        trainMat.append(setOfWords2Vec(vocabList,docList[docIndex]))#将训练数据变成单词向量
        trainClasses.append(classList[docIndex])#保存该训练集的类型
    p0V,p1V,pSpam=trainNBO(array(trainMat),array(trainClasses))#生成每个位置的权重
    #对测试集分类
    errorCount=0
    for docIndex in testSet:
        wordVector=setOfWords2Vec(vocabList,docList[docIndex])
        if classifyNB(array(wordVector),p0V,p1V,pSpam)!=classList[docIndex]:#分类结果如果不一样error+1
            errorCount+=1
    print("the error rate is :",float(errorCount)/len(testSet))

spamTest()