代码注释：机器学习实战第4章 基于概率论的分类方法：朴素贝叶斯

写在开头的话：在学习《机器学习实战》的过程中发现书中很多代码并没有注释，这对新入门的同学是一个挑战，特此贴出我对代码做出的注释，仅供参考，欢迎指正。

1、进行文本分类

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] #ddf
return postingList, classVec

#功能：创建不重复词的列表
#输入：数据集
#输出：不重复词的列表
def createVocabList(dataSet):
vocabSet = set([])#创建空集合
for document in dataSet:
vocabSet = vocabSet | set(document)#或操作表示创建并集
return list(vocabSet)

#功能：输入文档中的单词在词汇表中是否出现
#输入：词汇表，输入文档
#输出：文档向量
def setOfWords2Vec(vocabList, inputSet):
returnVec = [0] * len(vocabList)#创建和词汇表等长的向量
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

#功能：朴素贝叶斯分类器训练函数
#输入：文档向量，标签向量
#输出：任意文档属于侮辱性文档的概率，词汇在侮辱性文档中出现概率，词汇在非侮辱性文档中出现概率
def trainNB0(trainMatrix, trainCategory):
numTrainDocs = len(trainMatrix)#trainMatrix行数
numWords = len(trainMatrix[0])#trainMatrix列数
pAbusive = sum(trainCategory) / float(numTrainDocs)#任意文档属于侮辱性文档的概率
p0Num = ones(numWords)#拉普拉斯修正
p1Num = ones(numWords)#拉普拉斯修正
p0Denom = 2.0#拉普拉斯修正，只有两个标签
p1Denom = 2.0#拉普拉斯修正，只有两个标签
for i in range(numTrainDocs):
if trainCategory[i] == 1:#属于侮辱性文档
p1Num += trainMatrix[1]#侮辱性文档数量
p1Denom += sum(trainMatrix[i])
else:#不属于侮辱性文档
p0Num += trainMatrix[i]#非侮辱性文档数量
p0Denom += sum(trainMatrix[i])
p1Vect = log(p1Num / p1Denom)#防止下溢出，自然对数处理
p0Vect = log(p0Num / p0Denom)#防止下溢出，自然对数处理
return p0Vect, p1Vect, pAbusive

#功能：判断属于哪一类
#输入：要判断向量，词汇在侮辱性文档中出现概率，词汇在非侮辱性文档中出现概率，任意文档属于侮辱性文档的概率
#输出：判断值
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

#功能：输入文档中的单词在词汇表中是否出现
#输入：词汇表，输入文档
#输出：文档向量
def setOfWords2Vec(vocabList, inputSet):
returnVec = [0] * len(vocabList) # 创建和词汇表等长的向量
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 bagOfWords2VecMN(vocabList, inputSet):
returnVec = [0]*len(vocabList)
for word in inputSet:
if word in vocabList:
returnVec[vocabList.index(word)] += 1
return returnVec

#功能：便利函数
#输入：无
#输出：无
def testingNB():
label = ['ham', 'spam']#将结果以文本形式呈现
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: ', label[classifyNB(thisDoc, p0V, p1V, pAb)]
testEntry = ['stupid', 'garbage']
thisDoc = array(setOfWords2Vec(myVocabList, testEntry))
print testEntry, 'classified as: ', label[classifyNB(thisDoc, p0V, p1V, pAb)]

2、过滤垃圾邮件

#功能：切分文本
#输入：需要切分文本
#输出：切分好的文本
def textParse(bigString):
import re
listOfTokens = re.split(r'\W*', bigString)#\W表示非单词字符，*表示匹配前一个字符0次或无限次
#参考链接：http://www.cnblogs.com/huxi/archive/2010/07/04/1771073.html
return [tok.lower() for tok in listOfTokens if len(tok) > 2]#列表推导式

#功能：测试算法
#输入：无
#输出：测试正确率
def spamTest():
docList = []#文档列表
classList = []#标签列表
fullText = []#单词列表
for i in range(1, 26):
wordList = textParse(open('email/spam/%d.txt' % i).read())#打开spam文档，切分文本
docList.append(wordList)#append向列表尾部添加一个新的元素，将整个wordList添加进去，即文本为基本单位
fullText.extend(wordList)#extend向列表尾部添加wordList的所有元素，即单词为基本单位
classList.append(1)#标签表示为spam
wordList = textParse(open('email/ham/%d.txt' % i).read()) # 打开ham文档，切分文本
docList.append(wordList) # append向列表尾部添加一个新的元素，将整个wordList添加进去，即文本为基本单位
fullText.extend(wordList) # extend向列表尾部添加wordList的所有元素，即单词为基本单位
classList.append(0)#标签表示为ham
vocabList = createVocabList(docList)#创建不重复词的列表
trainingSet = range(50)
testSet = []
for i in range(10):#在trainingSet中随机删除十个数，删除的作为测试集，没有删除的作为训练集
randIndex = int(random.uniform(0, len(trainingSet)))
testSet.append(trainingSet[randIndex])
del(trainingSet[randIndex])
trainMat = []
trainClasses = []
for docIndex in trainingSet:#得测试集和测试标签
trainMat.append(bagOfWords2VecMN(vocabList, docList[docIndex]))
trainClasses.append(classList[docIndex])
p0V, p1V, pSpam = trainNB0(array(trainMat), array(trainClasses))
errorCount = 0
for docIndex in testSet:
wordVector = bagOfWords2VecMN(vocabList, docList[docIndex])
if classifyNB(array(wordVector), p0V, p1V, pSpam) != classList[docIndex]:#判断测试是否正确
errorCount += 1
print "classifiction error", docList[docIndex]
print 'the error rate is: ', float(errorCount) / len(testSet)

3、RSS应用

#功能：得出现频率最高的30个词
#输入：词汇表，文本
#输出：频率最高的30个词出现的个数和词
def calcMostFreq(vocabLi
4000
st, fullText):
import operator
freqDict = {}#建立字典
for token in vocabList:
freqDict[token] = fullText.count(token)
sortedFreq = sorted(freqDict.iteritems(), key = operator.itemgetter(1), reverse = True)#按值从小到大的顺序排序
return sortedFreq[:30]#返回前30 单词出现个数和词

#功能：rss功能
#输入：
#输出：
def localWords(feed1, feed0):
import feedparser
docList = []
classList = []
fullText = []
minLen = min(len(feed1['entries']), len(feed0['entries']))
for i in range(minLen):
wordList = textParse(feed1['entries'][i]['summary'])
docList.append(wordList)
fullText.extend(wordList)
classList.append(1)
wordList = textParse(feed0['entries'][i]['summary'])
docList.append(wordList)
fullText.extend(wordList)
classList.append(0)
vocabList = createVocabList(docList)
top30Words = calcMostFreq(vocabList, fullText)
for pairW in top30Words:
if pairW[0] in vocabList:
vocabList.remove(pairW[0])
trainingSet = range(2 * minLen)
testSet = []
for i in range(20):
randIndex = int(random.uniform(0, len(trainingSet)))
testSet.append(trainingSet[randIndex])
del(trainingSet[randIndex])
trainMat = []
trainClasses = []
for docIndex in trainingSet:
trainMat.append(bagOfWords2VecMN(vocabList, docList[docIndex]))
trainClasses.append(classList[docIndex])
p0V, p1V, pSpam = trainNB0(array(trainMat), array(trainClasses))
errorCount = 0
for docIndex in testSet:
wordVector = bagOfWords2VecMN(vocabList, docList[docIndex])
if classifyNB(array(wordVector), p0V, p1V, pSpam) != classList[docIndex]:
errorCount += 1
print 'the error rate is: ', float(errorCount) / len(testSet)
return vocabList, p0V, p1V

def getTopWords(nf, sf):
import operator
vocabList, p0V, p1V = localWords(nf, sf)
topNY = []
topSF = []
for i in range(len(p0V)):
if p0V[i] > -6.0 : topSF.append((vocabList[i], p0V[i]))
if p1V[i] > -6.0 : topNY.append((vocabList[i], p1V[i]))
sortedSF = sorted(topSF, key = lambda pair : pair[1], reverse = True)
print "SF**SF**SF**SF**SF**SF**SF**SF**SF**SF**SF**SF**SF**SF**SF**SF**"
for item in sortedSF:
print item[0]
sortedNY = sorted(topNY, key=lambda pair: pair[1], reverse=True)
print "NY**NY**NY**NY**NY**NY**NY**NY**NY**NY**NY**NY**NY**NY**NY**NY**"
for item in sortedNY:
print item[0]