使用python3 学习朴素贝叶斯分类api

设计到字符串提取特征向量

欢迎来到我的git下载源代码: https://github.com/linyi0604/MachineLearning

 

1 from sklearn.datasets import fetch_20newsgroups 2 from sklearn.cross_validation import train_test_split 3 # 导入文本特征向量转化模块 4 from sklearn.feature_extraction.text import CountVectorizer 5 # 导入朴素贝叶斯模型 6 from sklearn.naive_bayes import MultinomialNB 7 # 模型评估模块 8 from sklearn.metrics import classification_report 9 10 '''11 朴素贝叶斯模型广泛用于海量互联网文本分类任务。12 由于假设特征条件相互独立，预测需要估计的参数规模从幂指数量级下降接近线性量级，节约内存和计算时间13 但是 该模型无法将特征之间的联系考虑，数据关联较强的分类任务表现不好。14 '''15 16 '''17 1 读取数据部分18 '''19 # 该api会即使联网下载数据20 news = fetch_20newsgroups(subset="all")21 # 检查数据规模和细节22 # print(len(news.data))23 # print(news.data[0])24 '''25 1884626 27 From: Mamatha Devineni Ratnam 
     
      28 Subject: Pens fans reactions29 Organization: Post Office, Carnegie Mellon, Pittsburgh, PA30 Lines: 1231 NNTP-Posting-Host: po4.andrew.cmu.edu32 33 I am sure some bashers of Pens fans are pretty confused about the lack34 of any kind of posts about the recent Pens massacre of the Devils. Actually,35 I am  bit puzzled too and a bit relieved. However, I am going to put an end36 to non-PIttsburghers' relief with a bit of praise for the Pens. Man, they37 are killing those Devils worse than I thought. Jagr just showed you why38 he is much better than his regular season stats. He is also a lot39 fo fun to watch in the playoffs. Bowman should let JAgr have a lot of40 fun in the next couple of games since the Pens are going to beat the pulp out of Jersey anyway. I was very disappointed not to see the Islanders lose the final41 regular season game.          PENS RULE!!!42 '''43 44 '''45 2 分割数据部分46 '''47 x_train, x_test, y_train, y_test = train_test_split(news.data,48                                                     news.target,49                                                     test_size=0.25,50                                                     random_state=33)51 52 '''53 3 贝叶斯分类器对新闻进行预测54 '''55 # 进行文本转化为特征56 vec = CountVectorizer()57 x_train = vec.fit_transform(x_train)58 x_test = vec.transform(x_test)59 # 初始化朴素贝叶斯模型60 mnb = MultinomialNB()61 # 训练集合上进行训练， 估计参数62 mnb.fit(x_train, y_train)63 # 对测试集合进行预测 保存预测结果64 y_predict = mnb.predict(x_test)65 66 '''67 4 模型评估68 '''69 print("准确率:", mnb.score(x_test, y_test))70 print("其他指标：\n",classification_report(y_test, y_predict, target_names=news.target_names))71 '''72 准确率: 0.839770797962648573 其他指标：74                            precision    recall  f1-score   support75 76              alt.atheism       0.86      0.86      0.86       20177            comp.graphics       0.59      0.86      0.70       25078  comp.os.ms-windows.misc       0.89      0.10      0.17       24879 comp.sys.ibm.pc.hardware       0.60      0.88      0.72       24080    comp.sys.mac.hardware       0.93      0.78      0.85       24281           comp.windows.x       0.82      0.84      0.83       26382             misc.forsale       0.91      0.70      0.79       25783                rec.autos       0.89      0.89      0.89       23884          rec.motorcycles       0.98      0.92      0.95       27685       rec.sport.baseball       0.98      0.91      0.95       25186         rec.sport.hockey       0.93      0.99      0.96       23387                sci.crypt       0.86      0.98      0.91       23888          sci.electronics       0.85      0.88      0.86       24989                  sci.med       0.92      0.94      0.93       24590                sci.space       0.89      0.96      0.92       22191   soc.religion.christian       0.78      0.96      0.86       23292       talk.politics.guns       0.88      0.96      0.92       25193    talk.politics.mideast       0.90      0.98      0.94       23194       talk.politics.misc       0.79      0.89      0.84       18895       talk.religion.misc       0.93      0.44      0.60       15896 97              avg / total       0.86      0.84      0.82      471298 '''