机器学习实战（三）朴素贝叶斯分类

朴素贝叶斯分类

一、优缺点

1. 优点：在数据较少的时候仍然有效，可以处理多类别问题
2. 缺点：对于输入数据的准备方式较为敏感
3. 适用数据类型：标称型数据

二、朴素贝叶斯的一般流程

1. 收集数据
2. 准备数据
3. 分析数据
4. 训练算法
5. 测试算法
6. 使用算法

三、代码实例

训练数据
链接: https://pan.baidu.com/s/1s4EubJm38qAUhe6PPieQkA 提取码: abj2

# -*- coding: utf-8 -*-
# @Time    : 19-4-4 上午9:43
# @Author  : MRB
# @File    : email_classify.py
# @Software: PyCharm Community Edition

# import re
import sys,os
from numpy import *

#---------------------------词表到向量的转换函数--------------------------------------------------------
#创建实验样本
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

#创建一个包含所有文档中出现的不重复的列表
def createVocabList(dataSet):
    vocabSet = set([]) #创建一个空集
    for document in dataSet:
        vocabSet = vocabSet|set(document)  #两个集合的并集
    return list(vocabSet)

#输出文档向量
def setOfWord2Vec(vocabList,inputSet):
    returnVec = [0]*len(vocabList)
    for word in inputSet:
        if word in vocabList:
            returnVec[vocabList.index(word)] = 1
        else:
            print("the word:{} is not in my Vocabulary!".format(word))
    return returnVec

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

#---------------------------词表到向量的转换函数 end ------------------------------------------------------

'''
朴素贝叶斯分类器训练算法
'''
def trainNB0(trainMatrix,trainCategory):
    '''
    :param trainMatrix: 文档矩阵
    :param trainCategory: 类别标签构成的向量
    :return:
    '''
    numTrainDocs = len(trainMatrix)
    numWords = len(trainMatrix[0])

    pAbusive = sum(trainCategory)/float(numTrainDocs) #计算侮辱类的概率--3/6

    # 初始化概率
    # p0Num = zeros(numWords)
    # p1Num = zeros(numWords)
    # p0Denom = 0.0
    # p1Denom = 0.0
    #优化
    '''
    防止一个概率为0，相乘是结果为0,改为对数相加防止这种
    '''
    p0Num = ones(numWords)
    p1Num = ones(numWords)
    p0Denom = 2.0
    p1Denom = 2.0

    #向量相加
    for i in range(numTrainDocs):
        # print(trainMatrix[i])
        if trainCategory[i] == 1:
            p1Num += trainMatrix[i]
            p1Denom += sum(trainMatrix[i])
        else:
            p0Num += trainMatrix[i]
            p0Denom += sum(trainMatrix[i])
    #对每个元素做除法
    # p1Vect = p1Num/p1Denom #计算每个词条在侮辱类中出现的概率
    # p0Vect = p0Num/p0Denom #计算每个词条在非侮辱类中出现的概率
    #优化
    p1Vect = log(p1Num / p1Denom)  # 计算每个词条在侮辱类中出现的概率
    p0Vect = log(p0Num / p0Denom)  # 计算每个词条在非侮辱类中出现的概率
    return p0Vect,p1Vect,pAbusive

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

def testingNB():
    listOPlsts,listClasses = loadDataSet()
    myVocablist = createVocabList(listOPlsts)
    trainMat = []
    for  postinDoc in listOPlsts:
        trainMat.append(setOfWord2Vec(myVocablist,postinDoc))
    p0v,p1v,pAb = trainNB0(trainMat,listClasses)

    testEntry =['stupid','garbage']
    thisDoc = array(setOfWord2Vec(myVocablist,testEntry))
    print("{} classified as : {}".format(testEntry,classifyNB(thisDoc,p0v,p1v,pAb)))

    testEntry = ['love','my','dalmation']
    thisDoc = array(setOfWord2Vec(myVocablist, testEntry))
    print("{} classified as : {}".format(testEntry, classifyNB(thisDoc, p0v, p1v, pAb)))



'''
文件解析以及完整的来及邮件分类
'''

def textParse(bigString):
    import re
    listOfTokens = re.split(r'\W*',bigString)
    # print(listOfTokens)
    # print([tok for tok in listOfTokens if len(tok)>2])
    return [tok for tok in listOfTokens if len(tok)>2]

def spamTest():
    docList = []
    classList = []
    fulltext = []

    #导入解析文本文件
    for i in range(1,26):
        # print(i)
        wordList = textParse(open('email/spam/{}.txt'.format(str(i))).read())
        docList.append(wordList)
        fulltext.extend(wordList)
        classList.append(1)
        wordList = textParse(open('email/ham/{}.txt'.format(str(i))).read())
        docList.append(wordList)
        fulltext.extend(wordList)
        classList.append(0)
    vocabList = createVocabList(docList)
    trainingSet = list(range(50))
    testSet = []
    #随机构建训练集
    for i in range(10):
        randIndex = int(random.uniform(0,len(trainingSet)))
        testSet.append(trainingSet[randIndex])
        print(trainingSet)
        del(trainingSet[randIndex])
        # print(trainingSet)
    trainMat = []
    trainClasses = []
    for docIndex in trainingSet:
        trainMat.append(setOfWord2Vec(vocabList,docList[docIndex]))
        trainClasses.append(classList[docIndex])
    p0V,p1V,pSpam = trainNB0(array(trainMat),array(trainClasses))
    errorCount = 0
    for docIndex in testSet:
        wordVector = setOfWord2Vec(vocabList,docList[docIndex])
        if classifyNB(array(wordVector),p0V,p1V,pSpam) != classList[docIndex]:
            errorCount += 1
    print("the error rate is : ",float(errorCount)/len(testSet))

if __name__ == '__main__':
    spamTest()