python实现朴素贝叶斯算法

时间:2022-11-13 20:08:32

本代码实现了朴素贝叶斯分类器(假设了条件独立的版本),常用于垃圾邮件分类,进行了拉普拉斯平滑。

关于朴素贝叶斯算法原理可以参考博客中原理部分的博文。

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
#!/usr/bin/python
# -*- coding: utf-8 -*-
from math import log
from numpy import*
import operator
import matplotlib
import matplotlib.pyplot as plt
from os import listdir
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
def createVocabList(dataSet):
  vocabSet = set([]) #create empty set
  for document in dataSet:
    vocabSet = vocabSet | set(document) #union of the two sets
  return list(vocabSet)
 
def setOfWords2Vec(vocabList, inputSet):
  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):  #训练模型
  numTrainDocs = len(trainMatrix)
  numWords = len(trainMatrix[0])
  pAbusive = sum(trainCategory)/float(numTrainDocs)
  p0Num = ones(numWords); p1Num = ones(numWords)  #拉普拉斯平滑
  p0Denom = 0.0+2.0; p1Denom = 0.0 +2.0      #拉普拉斯平滑
  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)    #用log()是为了避免概率乘积时浮点数下溢
  p0Vect = log(p0Num/p0Denom)
  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 bagOfWords2VecMN(vocabList, inputSet):
  returnVec = [0] * len(vocabList)
  for word in inputSet:
    if word in vocabList:
      returnVec[vocabList.index(word)] += 1
  return returnVec
 
def testingNB():  #测试训练结果
  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)
 
def textParse(bigString): # 长字符转转单词列表
  import re
  listOfTokens = re.split(r'\W*', bigString)
  return [tok.lower() for tok in listOfTokens if len(tok) > 2]
 
def spamTest():  #测试垃圾文件 需要数据
  docList = [];
  classList = [];
  fullText = []
  for i in range(1, 26):
    wordList = textParse(open('email/spam/%d.txt' % i).read())
    docList.append(wordList)
    fullText.extend(wordList)
    classList.append(1)
    wordList = textParse(open('email/ham/%d.txt' % i).read())
    docList.append(wordList)
    fullText.extend(wordList)
    classList.append(0)
  vocabList = createVocabList(docList)
  trainingSet = range(50);
  testSet = []
  for i in range(10):
    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 "classification error", docList[docIndex]
  print 'the error rate is: ', float(errorCount) / len(testSet)
 
 
 
listOPosts,listClasses=loadDataSet()
myVocabList=createVocabList(listOPosts)
print myVocabList,'\n'
# print setOfWords2Vec(myVocabList,listOPosts[0]),'\n'
trainMat=[]
for postinDoc in listOPosts:
  trainMat.append(setOfWords2Vec(myVocabList,postinDoc))
print trainMat
p0V,p1V,pAb=trainNB0(trainMat,listClasses)
print pAb
print p0V,'\n',p1V
testingNB()

以上就是本文的全部内容,希望对大家的学习有所帮助,也希望大家多多支持服务器之家。

原文链接:https://blog.csdn.net/qq_35083093/article/details/79107514