代碼來源于：https://www.cnblogs.com/huangyc/p/10327209.html? ，本人只是簡(jiǎn)介學(xué)習(xí)

1、?貝葉斯.py

import numpy as np from word_utils import *class NaiveBayesBase(object):def __init__(self):passdef fit(self, trainMatrix, trainCategory):'''樸素貝葉斯分類器訓(xùn)練函數(shù)，求：p(Ci),基于詞匯表的p(w|Ci)Args:trainMatrix : 訓(xùn)練矩陣，即向量化表示后的文檔（詞條集合）trainCategory : 文檔中每個(gè)詞條的列表標(biāo)注Return:p0Vect : 屬于0類別的概率向量(p(w1|C0),p(w2|C0),...,p(wn|C0))p1Vect : 屬于1類別的概率向量(p(w1|C1),p(w2|C1),...,p(wn|C1))pAbusive : 屬于1類別文檔的概率'''numTrainDocs = len(trainMatrix)# 長(zhǎng)度為詞匯表長(zhǎng)度numWords = len(trainMatrix[0])# p(ci)self.pAbusive = sum(trainCategory) / float(numTrainDocs)# 由于后期要計(jì)算p(w|Ci)=p(w1|Ci)*p(w2|Ci)*...*p(wn|Ci)，若wj未出現(xiàn)，則p(wj|Ci)=0,因此p(w|Ci)=0，這樣顯然是不對(duì)的# 故在初始化時(shí)，將所有詞的出現(xiàn)數(shù)初始化為1，分母即出現(xiàn)詞條總數(shù)初始化為2p0Num = np.ones(numWords)p1Num = np.ones(numWords)p0Denom = 2.0p1Denom = 2.0for i in range(numTrainDocs):if trainCategory[i] == 1:p1Num += trainMatrix[i]p1Denom += sum(trainMatrix[i])else:p0Num += trainMatrix[i]p0Denom += sum(trainMatrix[i])# p(wi | c1)# 為了避免下溢出（當(dāng)所有的p都很小時(shí)，再相乘會(huì)得到0.0，使用log則會(huì)避免得到0.0）self.p1Vect = np.log(p1Num / p1Denom)# p(wi | c2)self.p0Vect = np.log(p0Num / p0Denom)return selfdef predict(self, testX):'''樸素貝葉斯分類器Args:testX : 待分類的文檔向量（已轉(zhuǎn)換成array）p0Vect : p(w|C0)p1Vect : p(w|C1)pAbusive : p(C1)Return:1 : 為侮辱性文檔 (基于當(dāng)前文檔的p(w|C1)*p(C1)=log(基于當(dāng)前文檔的p(w|C1))+log(p(C1)))0 : 非侮辱性文檔 (基于當(dāng)前文檔的p(w|C0)*p(C0)=log(基于當(dāng)前文檔的p(w|C0))+log(p(C0)))'''p1 = np.sum(testX * self.p1Vect) + np.log(self.pAbusive)p0 = np.sum(testX * self.p0Vect) + np.log(1 - self.pAbusive)if p1 > p0:return 1else:return 0def loadDataSet():'''數(shù)據(jù)加載函數(shù)。這里是一個(gè)小例子'''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代表侮辱性文字，0代表正常言論，代表上面6個(gè)樣本的類別return postingList, classVecdef checkNB():'''測(cè)試'''listPosts, listClasses = loadDataSet()myVocabList = createVocabList(listPosts)trainMat = []for postDoc in listPosts:trainMat.append(setOfWord2Vec(myVocabList, postDoc))nb = NaiveBayesBase()nb.fit(np.array(trainMat), np.array(listClasses))testEntry1 = ['love', 'my', 'dalmation']thisDoc = np.array(setOfWord2Vec(myVocabList, testEntry1))print(testEntry1, 'classified as:', nb.predict(thisDoc))testEntry2 = ['stupid', 'garbage']thisDoc2 = np.array(setOfWord2Vec(myVocabList, testEntry2))print(testEntry2, 'classified as:', nb.predict(thisDoc2))if __name__ == "__main__":checkNB() View Code

2、word_utils.py

def createVocabList(dataSet):'''創(chuàng)建所有文檔中出現(xiàn)的不重復(fù)詞匯列表Args:dataSet: 所有文檔Return:包含所有文檔的不重復(fù)詞列表，即詞匯表'''vocabSet = set([])# 創(chuàng)建兩個(gè)集合的并集for document in dataSet:vocabSet = vocabSet | set(document)return list(vocabSet)# 詞袋模型(bag-of-words model):詞在文檔中出現(xiàn)的次數(shù) def bagOfWords2Vec(vocabList, inputSet):'''依據(jù)詞匯表，將輸入文本轉(zhuǎn)化成詞袋模型詞向量Args:vocabList: 詞匯表inputSet: 當(dāng)前輸入文檔Return:returnVec: 轉(zhuǎn)換成詞向量的文檔例子：vocabList = ['I', 'love', 'python', 'and', 'machine', 'learning']inputset = ['python', 'machine', 'learning', 'python', 'machine']returnVec = [0, 0, 2, 0, 2, 1]長(zhǎng)度與詞匯表一樣長(zhǎng)，出現(xiàn)了的位置為1，未出現(xiàn)為0，如果詞匯表中無該單詞則print'''returnVec = [0] * len(vocabList)for word in inputSet:if word in vocabList:returnVec[vocabList.index(word)] += 1else:print("the word: %s is not in my vocabulary!" % word)return returnVec# 詞集模型(set-of-words model):詞在文檔中是否存在，存在為1，不存在為0 def setOfWord2Vec(vocabList, inputSet):'''依據(jù)詞匯表，將輸入文本轉(zhuǎn)化成詞集模型詞向量Args:vocabList: 詞匯表inputSet: 當(dāng)前輸入文檔Return:returnVec: 轉(zhuǎn)換成詞向量的文檔例子：vocabList = ['I', 'love', 'python', 'and', 'machine', 'learning']inputset = ['python', 'machine', 'learning']returnVec = [0, 0, 1, 0, 1, 1]長(zhǎng)度與詞匯表一樣長(zhǎng)，出現(xiàn)了的位置為1，未出現(xiàn)為0，如果詞匯表中無該單詞則print'''returnVec = [0] * len(vocabList)for word in inputSet:if word in vocabList:returnVec[vocabList.index(word)] = 1else:print("the word: %s is not in my vocabulary!" % word)return returnVec View Code

?

轉(zhuǎn)載于:https://www.cnblogs.com/ywjfx/p/11045395.html

總結(jié)

以上是生活随笔為你收集整理的机器学习之贝叶斯垃圾邮件分类的全部?jī)?nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。