前言

代码实现

使用朴素贝叶斯进行文档分类

import numpy as np


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


# 创建并集
def createVocabList(dataSet):
    vocabSet = set([])
    for document in dataSet:
        vocabSet = vocabSet | set(document)
    return list(vocabSet)


# inputSet：输入的文档；vocabList：词汇表；该方法用于查看词汇表中的词汇是否在输入的文档中出现
def setOfWord2Vec(vocabList, inputSet):
    # 创建和词汇表等长的向量
    returnVec = [0]*len(vocabList)
    for word in inputSet:
        # 如果输入文档中包含了侮辱词汇表中的单词，就把输出文档想两种的对应值设置为1
        if word in vocabList:
            returnVec[vocabList.index(word)] = 1
        else:
            print("The word: %s is not in my vocabulary!" % word)
    return returnVec


# 利用贝叶斯计算文档属于每个类别的概率
# trainMatrix：输入的文档矩阵
def trainNB0(trainMatrix, trainCategory):
    # 获取文档数量
    numTrainDocs = len(trainMatrix)
    # 获取文档中的词汇数量
    numWords = len(trainMatrix[0])
    pAbusive = sum(trainCategory) / float(numTrainDocs)
    # 初始化概率：为防止概率为0，分子初始化为1，分母初始化为2
    p0Num = np.ones(numWords)
    p1Num = np.ones(numWords)
    p0Denom = 2.0
    p1Denom = 2.0
    # p0Num = np.zeros(numWords)
    # p1Num = np.zeros(numWords)
    # p0Denom = 0.0
    # p1Denom = 0.0
    for i in range(numTrainDocs):
        if trainCategory[i] == 1:
            p1Num += trainMatrix[i]
            p1Denom += sum(trainMatrix[i])
        else:
            p0Num += trainMatrix[i]
            p0Denom += sum(trainMatrix[i])
    # 为了避免在Python中下溢出，将乘积处理转换为对数处理
    p1Vect = np.log(p1Num / p1Denom)
    p0Vect = np.log(p0Num / p0Denom)
    # p1Vect = p1Num / p1Denom
    # p0Vect = p0Num / p0Denom
    return p0Vect, p1Vect, pAbusive


def classifyNB(vec2Classify, p0Vec, p1Vec, pClass1):
    """
    :param vec2Classify:需要分类的向量
    :param p0Vec:分类为0的条件概率
    :param p1Vec:分类为1的条件概率
    :param pClass1:文档属于分类1的概率
    :return:
    """
    # 计算
    p1 = sum(vec2Classify * p1Vec) + np.log(pClass1)
    p0 = sum(vec2Classify * p0Vec) + np.log(1.0 - pClass1)
    if p1 > p0:
        return 1
    else:
        return 0


if __name__ == "__main__":
    listPosts, listClasses = loadDataSet()
    myVocabList = createVocabList(listPosts)
    # print(myVocabList)
    # print(setOfWord2Vec(myVocabList, listPosts[0]))

    trainMat = []
    for postInDoc in listPosts:
        trainMat.append(setOfWord2Vec(myVocabList, postInDoc))      # 计算每个文档中词表词汇出现次数，并加入文档矩阵中去
    p0V, p1V, pAb = trainNB0(np.array(trainMat), np.array(listClasses))
    testEntry = ['love', 'my', 'dalmation']
    thisDoc = np.array(setOfWord2Vec(myVocabList, testEntry))
    print(testEntry, "classified as:", classifyNB(thisDoc, p0V, p1V, pAb))
    testEntry = ['stupid', 'garbage']
    thisDoc = np.array(setOfWord2Vec(myVocabList, testEntry))
    print(testEntry, "classified as:", classifyNB(thisDoc, p0V, p1V, pAb))