机器学习实战 朴素贝叶斯

#!/usr/bin/env python
# -*- coding: utf-8 -*-

from numpy import *

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 is abusive, 0 not
    return postingList,classVec
                 
def createVocabList(dataSet):
    vocabSet = set([])  #create empty set
    for document in dataSet:
        set_document = set(document)
        vocabSet = vocabSet | set_document #union of the two sets
        list_vocabSet = list(vocabSet)
    return list_vocabSet

def setOfWords2Vec(vocabList, inputSet):#词表到向量表的转换
    returnVec = [0]*len(vocabList)
    for word in inputSet:
        if word in vocabList:
            returnVec[vocabList.index(word)] = 1
        else: print "the word: %s is not in my Vocabulary!" % word
    return returnVec

def trainNB0(trainMatrix,trainCategory):
    numTrainDocs = len(trainMatrix)
    numWords = len(trainMatrix[0])
    sum_trainCategory = sum(trainCategory)
    float_numTrainDocs = float(numTrainDocs)
    pAbusive = sum_trainCategory / float_numTrainDocs
#    pAbusive = sum(trainCategory)/float(numTrainDocs)
    p0Num = ones(numWords); p1Num = ones(numWords)      #change to ones()
    p0Denom = 2.0; p1Denom = 2.0                        #change to 2.0  #分子初始化为1 分母初始化为2  避免P(W|C)=0
    for i in range(numTrainDocs):
        if trainCategory[i] == 1:
            p1Num += trainMatrix[i]
            p1Denom += sum(trainMatrix[i])
        else:
            p0Num += trainMatrix[i]
            p0Denom += sum(trainMatrix[i])
    p1Vect = log(p1Num/p1Denom)          #change to log()  #log[P(W|C1)]
    p0Vect = log(p0Num/p0Denom)          #change to log()
    return p0Vect,p1Vect,pAbusive

# P(C1|W) = P(W|C1)*P(C1)/P(W)    P(C2|W) = P(W|C2)*P(C2)/P(W)
# P(W)相同  比较P(W|C1)*P(C1)和P(W|C2)*P(C2)大小
# ln(a*b)=ln(a)+ln(b) ln避免很多小的数相乘出现下溢
#log[P(W|C1)*P(C1)] = log[P(W|C1)]+P(C1)  即 p1 = sum(vec2Classify * p1Vec) + log(pClass1)
#log[P(W|C1)]    即 p1Vect = log(p1Num/p1Denom)

def classifyNB(vec2Classify, p0Vec, p1Vec, pClass1):
    p1 = sum(vec2Classify * p1Vec) + log(pClass1)    #log[P(W|C1)*P(C1)] = log[P(W|C1)]+P(C1)
    p0 = sum(vec2Classify * p0Vec) + log(1.0 - pClass1)
    if p1 > p0:
        return 1
    else: 
        return 0

def bagOfWords2VecMN(vocabList, inputSet):
    returnVec = [0]*len(vocabList)
    for word in inputSet:
        if word in vocabList:
            returnVec[vocabList.index(word)] += 1
    return returnVec

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

if __name__ == '__main__':
    listOPosts, listClasses = loadDataSet()
    myVocabList = createVocabList(listOPosts)
    a = setOfWords2Vec(myVocabList, listOPosts[0])
    b = setOfWords2Vec(myVocabList, listOPosts[3])
    print 'myVocabList:', myVocabList
    print 'a:', a
    print 'b:', b


    result = testingNB()