学习日志---朴素贝叶斯算法

import numpy as np

#返回样本数据集
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]
    return postingList,classVec

#提取样本数据中的单词，构成词汇表
def createVocabList(dataSet):
    vocabSet = set([])
    for document in dataSet:
        vocabSet = vocabSet | set(document)
    return list(vocabSet)

#传入单词表和待分析的数据，讲数据转为向量，这里记录每行样本的单词是否出现
def setOfWords2Vec(vocabList, inputSet):
    retVocabList = [0] * len(vocabList)
    for word in inputSet:
        if word in vocabList:
            retVocabList[vocabList.index(word)] = 1
        else:
            print ‘word ‘,word ,‘not in dict‘
    return retVocabList

#这里是每个样本的出现次数
def bagOfWords2VecMN(vocabList, inputSet):
    returnVec = [0]*len(vocabList)
    for word in inputSet:
        if word in vocabList:
            returnVec[vocabList.index(word)] += 1
    return returnVec

#带入样本数据和结果，计算样本对于某一类别的出现次数
#这个求出不同组中，每个词出现的概率
def trainNB0(trainMatrix,trainCatergory):
    numTrainDoc = len(trainMatrix)
    numWords = len(trainMatrix[0])
    pAbusive = sum(trainCatergory)/float(numTrainDoc)
    #防止多个概率的成绩当中的一个为0
    p0Num = np.ones(numWords)
    p1Num = np.ones(numWords)
    p0Denom = 2.0
    p1Denom = 2.0
    for i in range(numTrainDoc):
        if trainCatergory[i] == 1:
            p1Num +=trainMatrix[i]
            p1Denom += sum(trainMatrix[i])
        else:
            p0Num +=trainMatrix[i]
            p0Denom += sum(trainMatrix[i])
    #处于精度的考虑，否则很可能到限归零,因为可能有太多项都为0
    #避免下溢出和浮点数舍入导致的错误
    p1Vect = np.log(p1Num/p1Denom)
    p0Vect = np.log(p0Num/p0Denom)
    return p0Vect,p1Vect,pAbusive

#这里也就相当于log了一下
def classifyNB(vec2Classify, p0Vec, p1Vec, pClass1):
    p1 = sum(vec2Classify * p1Vec) + np.log(pClass1)
    p0 = sum(vec2Classify * p0Vec) + np.log(1.0 - pClass1)
    if p1 > p0:
        return 1
    else:
        return 0

#测试方法
def testingNB():
    listOPosts,listClasses = loadDataSet()
    myVocabList = createVocabList(listOPosts)
    trainMat=[]
    for postinDoc in listOPosts:
        trainMat.append(setOfWords2Vec(myVocabList, postinDoc))
    p0V,p1V,pAb = trainNB0(np.array(trainMat),np.array(listClasses))
    testEntry = [‘love‘, ‘my‘, ‘dalmation‘]
    thisDoc = np.array(setOfWords2Vec(myVocabList, testEntry))
    print testEntry,‘classified as: ‘,classifyNB(thisDoc,p0V,p1V,pAb)
    testEntry = [‘stupid‘, ‘garbage‘]
    thisDoc = np.array(setOfWords2Vec(myVocabList, testEntry))
    print testEntry,‘classified as: ‘,classifyNB(thisDoc,p0V,p1V,pAb)

def main():
    testingNB()

if __name__ == ‘__main__‘:
    main()