from sklearn.datasets import fetch_20newsgroups
news = fetch_20newsgroups()
from sklearn.model_selection import train_test_split
x_train, x_test, y_train, y_test = train_test_split(news.data, news.target, test_size=0.25, random_state=33)
from sklearn.feature_extraction.text import CountVectorizer
count_vec = CountVectorizer()
x_count_train = count_vec.fit_transform(x_train)
x_count_test = count_vec.transform(x_test)
from sklearn.naive_bayes import MultinomialNB
mnb_count = MultinomialNB()
mnb_count.fit(x_count_train, y_train)
print('The accuracy of classifying 20nesgroups using Naive Bayes(CountVectorizer without filtering stopwords):', mnb_count.score(x_count_test, y_test))
y_count_predict = mnb_count.predict(x_count_test)
from sklearn.metrics import classification_report
print(classification_report(y_test, y_count_predict, target_names=news.target_names))
运行结果如下:
The accuracy of classifying 20nesgroups using Naive Bayes(CountVectorizer without filtering stopwords): 0.831742665253
precision recall f1-score support
alt.atheism 0.89 0.88 0.88 108
comp.graphics 0.62 0.86 0.72 130
comp.os.ms-windows.misc 0.95 0.13 0.23 163
comp.sys.ibm.pc.hardware 0.56 0.77 0.64 141
comp.sys.mac.hardware 0.91 0.83 0.87 145
comp.windows.x 0.71 0.91 0.80 141
misc.forsale 0.95 0.67 0.78 159
rec.autos 0.87 0.90 0.89 139
rec.motorcycles 0.96 0.94 0.95 153
rec.sport.baseball 0.97 0.89 0.93 141
rec.sport.hockey 0.96 0.97 0.96 148
sci.crypt 0.75 0.99 0.85 143
sci.electronics 0.90 0.79 0.84 160
sci.med 0.95 0.89 0.92 158
sci.space 0.91 0.91 0.91 149
soc.religion.christian 0.79 0.97 0.87 157
talk.politics.guns 0.83 0.95 0.89 134
talk.politics.mideast 0.89 0.99 0.94 133
talk.politics.misc 0.78 0.91 0.84 130
talk.religion.misc 0.98 0.53 0.68 97