bayesSklearn.py

'''
Naive Bayes Source Code for chapter 4

@author: Teddy.Ma
'''
# hack for loading shared module even when current folder is a sub folder
import sys
if (sys.path[-1] != '..'): sys.path.append('..')

from shared.common import *
from numpy import *
import feedparser
from sklearn.naive_bayes import MultinomialNB,GaussianNB

# 4.5.1 generate test data
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

# 4.5.1 create a list of all the base words for training and classify
def createVocabList(dataSet):
    vocabSet = set([])  #create empty set
    for document in dataSet:
        vocabSet = vocabSet | set(document) #union of the two sets
    return list(vocabSet)

# 4.5.1 normalize a document to the 0/1 value base words based vector
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
    
# 4.5.3 test Naïve Bayes with sklearn MultinomialNB
def testingNB():
    listOPosts,listClasses = loadDataSet()
    myVocabList = createVocabList(listOPosts)
    trainMat=[]
    for postinDoc in listOPosts:
        trainMat.append(setOfWords2Vec(myVocabList, postinDoc))
    clf = MultinomialNB()
    clf.fit(trainMat, listClasses)
    testEntry = ['love', 'my', 'dalmation']
    thisDoc = array(setOfWords2Vec(myVocabList, testEntry))
    print (testEntry,'classified as: ',clf.predict([thisDoc])[0])
    testEntry = ['stupid', 'garbage']
    thisDoc = array(setOfWords2Vec(myVocabList, testEntry))
    print (testEntry,'classified as: ',clf.predict([thisDoc])[0])

# 4.5.3 test Naïve Bayes with sklearn GaussianNB
def testingNB2():
    listOPosts,listClasses = loadDataSet()
    myVocabList = createVocabList(listOPosts)
    trainMat=[]
    for postinDoc in listOPosts:
        trainMat.append(setOfWords2Vec(myVocabList, postinDoc))
    clf = GaussianNB()
    clf.fit(trainMat, listClasses)
    testEntry = ['love', 'my', 'dalmation']
    thisDoc = array(setOfWords2Vec(myVocabList, testEntry))
    print (testEntry,'classified as: ',clf.predict([thisDoc])[0])
    testEntry = ['stupid', 'garbage']
    thisDoc = array(setOfWords2Vec(myVocabList, testEntry))
    print (testEntry,'classified as: ',clf.predict([thisDoc])[0])

def textParse(bigString):    #input is big string, #output is word list
    import re
    listOfTokens = re.split(r'\W*', bigString)
    return [tok.lower() for tok in listOfTokens if len(tok) > 2] 

# 4.5.4 setOfWords2Vec to consider the weight of word occurence
def bagOfWords2VecMN(vocabList, inputSet):
    returnVec = [0]*len(vocabList)
    for word in inputSet:
        if word in vocabList:
            returnVec[vocabList.index(word)] += 1
    return returnVec
    
# 4.6.2 test spam email with sklearn MultinomialNB
def spamTest():
    docList=[]; classList = []; fullText =[]
    for i in range(1,26):
        wordList = textParse(open('email/spam/%d.txt' % i, 'r', encoding='Windows-1252').read())
        docList.append(wordList)
        fullText.extend(wordList)
        classList.append(1)
        wordList = textParse(open('email/ham/%d.txt' % i, 'r', encoding='Windows-1252').read())
        docList.append(wordList)
        fullText.extend(wordList)
        classList.append(0)
    vocabList = createVocabList(docList)#create vocabulary
    trainingSet = list(range(50)); testSet=[]           #create test set
    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:#train the classifier (get probs) trainNB0
        trainMat.append(bagOfWords2VecMN(vocabList, docList[docIndex]))
        trainClasses.append(classList[docIndex])
    clf = MultinomialNB()
    clf.fit(trainMat, trainClasses)
    errorCount = 0
    for docIndex in testSet:        #classify the remaining items
        wordVector = bagOfWords2VecMN(vocabList, docList[docIndex])
        if clf.predict([wordVector])[0] != classList[docIndex]:
            errorCount += 1
            print ("classification error",docList[docIndex])
    print ('the error rate is: ',float(errorCount)/len(testSet))
    #return vocabList,fullText

# 4.6.2 test spam email with sklearn GaussianNB
def spamTest2():
    docList=[]; classList = []; fullText =[]
    for i in range(1,26):
        wordList = textParse(open('email/spam/%d.txt' % i, 'r', encoding='Windows-1252').read())
        docList.append(wordList)
        fullText.extend(wordList)
        classList.append(1)
        wordList = textParse(open('email/ham/%d.txt' % i, 'r', encoding='Windows-1252').read())
        docList.append(wordList)
        fullText.extend(wordList)
        classList.append(0)
    vocabList = createVocabList(docList)#create vocabulary
    trainingSet = list(range(50)); testSet=[]           #create test set
    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:#train the classifier (get probs) trainNB0
        trainMat.append(bagOfWords2VecMN(vocabList, docList[docIndex]))
        trainClasses.append(classList[docIndex])
    clf = GaussianNB()
    clf.fit(trainMat, trainClasses)
    errorCount = 0
    for docIndex in testSet:        #classify the remaining items
        wordVector = bagOfWords2VecMN(vocabList, docList[docIndex])
        if clf.predict([wordVector])[0] != classList[docIndex]:
            errorCount += 1
            print ("classification error",docList[docIndex])
    print ('the error rate is: ',float(errorCount)/len(testSet))
    #return vocabList,fullText

# 4.7.1 only lick the top 30 most frequent words
def calcMostFreq(vocabList,fullText):
    freqDict = {}
    for token in vocabList:
        freqDict[token]=fullText.count(token)
    sortedFreq = dicSorted(freqDict.items()) 
    return sortedFreq[:30]       

# 4.7.1 compare local words of two feeds with sklearn MultinomialNB
def localWords(feed1,feed0):
    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) #NY is class 1
        wordList = textParse(feed0['entries'][i]['summary'])
        docList.append(wordList)
        fullText.extend(wordList)
        classList.append(0)
    vocabList = createVocabList(docList)#create vocabulary
    top30Words = calcMostFreq(vocabList,fullText)   #remove top 30 words
    for pairW in top30Words:
        if pairW[0] in vocabList: vocabList.remove(pairW[0])
    trainingSet = list(range(2*minLen)); testSet=[]   #create test set
    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:#train the classifier (get probs) trainNB0
        trainMat.append(bagOfWords2VecMN(vocabList, docList[docIndex]))
        trainClasses.append(classList[docIndex])
    clf = MultinomialNB()
    clf.fit(trainMat, trainClasses)
    errorCount = 0
    for docIndex in testSet:        #classify the remaining items
        wordVector = bagOfWords2VecMN(vocabList, docList[docIndex])
        if clf.predict([wordVector]) != classList[docIndex]:
            errorCount += 1
    print ('the error rate is: ',float(errorCount)/len(testSet))

# 4.7.1 compare local words of two feeds with sklearn GaussianNB
def localWords2(feed1,feed0):
    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) #NY is class 1
        wordList = textParse(feed0['entries'][i]['summary'])
        docList.append(wordList)
        fullText.extend(wordList)
        classList.append(0)
    vocabList = createVocabList(docList)#create vocabulary
    top30Words = calcMostFreq(vocabList,fullText)   #remove top 30 words
    for pairW in top30Words:
        if pairW[0] in vocabList: vocabList.remove(pairW[0])
    trainingSet = list(range(2*minLen)); testSet=[]   #create test set
    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:#train the classifier (get probs) trainNB0
        trainMat.append(bagOfWords2VecMN(vocabList, docList[docIndex]))
        trainClasses.append(classList[docIndex])
    clf = MultinomialNB()
    clf.fit(trainMat, trainClasses)
    errorCount = 0
    for docIndex in testSet:        #classify the remaining items
        wordVector = bagOfWords2VecMN(vocabList, docList[docIndex])
        if clf.predict([wordVector]) != classList[docIndex]:
            errorCount += 1
    print ('the error rate is: ',float(errorCount)/len(testSet))

# 4.7.1 & 4.7.2 test localWords with two feeds
# the original RSS feeds listed in the book no longer work,
# so using another two instead here
def testLocalWords():
    ny=feedparser.parse('https://www.bigblueview.com/rss')
    sf=feedparser.parse('http://www.7x7.com/feeds/feed.rss')
    localWords(ny,sf)
    localWords2(ny,sf)