Python を使用して小さなテキスト分類システムを実装する

import os
import pickle
from sklearn.datasets.base import Bunch
#Bunch 类提供了一种key，value的对象形式
#target_name 所有分类集的名称列表
#label 每个文件的分类标签列表
#filenames 文件路径
#contents 分词后文件词向量形式def readfile(path):
    fp = open(path, "r", encoding='gb2312', errors='ignore')
    content = fp.read()
    fp.close()
    return content
bunch=Bunch(target_name=[],label=[],filenames=[],contents=[])
# wordbag_path="train_word_bag/train_set.dat"
# seg_path="train_seg/"wordbag_path="test_word_bag/test_set.dat"seg_path="test_seg/"catelist=os.listdir(seg_path)
bunch.target_name.extend(catelist)#将类别信息保存到Bunch对象for mydir in catelist:
    class_path=seg_path+mydir+"/"
    file_list=os.listdir(class_path)
    for file_path in file_list:
        fullname=class_path+file_path
        bunch.label.append(mydir)#保存当前文件的分类标签
        bunch.filenames.append(fullname)#保存当前文件的文件路径
        bunch.contents.append(readfile(fullname).strip())#保存文件词向量
#Bunch对象持久化file_obj=open(wordbag_path,"wb")
pickle.dump(bunch,file_obj)
file_obj.close()
print("构建文本对象结束")

import os
from sklearn.datasets.base import Bunch
import pickle#持久化类from sklearn import feature_extraction
from sklearn.feature_extraction.text import TfidfTransformer#TF-IDF向量转换类from sklearn.feature_extraction.text import TfidfVectorizer#TF-IDF向量生成类def readbunchobj(path):
    file_obj=open(path,"rb")
    bunch=pickle.load(file_obj)
    file_obj.close()
    return bunch
def writebunchobj(path,bunchobj):
    file_obj=open(path,"wb")
    pickle.dump(bunchobj,file_obj)
    file_obj.close()
def readfile(path):
    fp = open(path, "r", encoding='gb2312', errors='ignore')
    content = fp.read()
    fp.close()
    return content
path="train_word_bag/train_set.dat"bunch=readbunchobj(path)
#停用词stopword_path="train_word_bag/hlt_stop_words.txt"stpwrdlst=readfile(stopword_path).splitlines()
#构建TF-IDF词向量空间对象tfidfspace=Bunch(target_name=bunch.target_name,label=bunch.label,filenames=bunch.filenames,tdm=[],vocabulary={})
#使用TfidVectorizer初始化向量空间模型vectorizer=TfidfVectorizer(stop_words=stpwrdlst,sublinear_tf=True,max_df=0.5)
transfoemer=TfidfTransformer()#该类会统计每个词语的TF-IDF权值
#文本转为词频矩阵，单独保存字典文件tfidfspace.tdm=vectorizer.fit_transform(bunch.contents)
tfidfspace.vocabulary=vectorizer.vocabulary_
#创建词袋的持久化space_path="train_word_bag/tfidfspace.dat"writebunchobj(space_path,tfidfspace)

import os
from sklearn.datasets.base import Bunch
import pickle#持久化类from sklearn import feature_extraction
from sklearn.feature_extraction.text import TfidfTransformer#TF-IDF向量转换类from sklearn.feature_extraction.text import TfidfVectorizer#TF-IDF向量生成类from TF_IDF import space_path
def readbunchobj(path):
    file_obj=open(path,"rb")
    bunch=pickle.load(file_obj)
    file_obj.close()
    return bunch
def writebunchobj(path,bunchobj):
    file_obj=open(path,"wb")
    pickle.dump(bunchobj,file_obj)
    file_obj.close()
def readfile(path):
    fp = open(path, "r", encoding='gb2312', errors='ignore')
    content = fp.read()
    fp.close()
    return content
#导入分词后的词向量bunch对象path="test_word_bag/test_set.dat"bunch=readbunchobj(path)
#停用词stopword_path="train_word_bag/hlt_stop_words.txt"stpwrdlst=readfile(stopword_path).splitlines()
#构建测试集TF-IDF向量空间testspace=Bunch(target_name=bunch.target_name,label=bunch.label,filenames=bunch.filenames,tdm=[],vocabulary={})
#导入训练集的词袋trainbunch=readbunchobj("train_word_bag/tfidfspace.dat")
#使用TfidfVectorizer初始化向量空间vectorizer=TfidfVectorizer(stop_words=stpwrdlst,sublinear_tf=True,max_df=0.5,vocabulary=trainbunch.vocabulary)
transformer=TfidfTransformer();
testspace.tdm=vectorizer.fit_transform(bunch.contents)
testspace.vocabulary=trainbunch.vocabulary
#创建词袋的持久化space_path="test_word_bag/testspace.dat"writebunchobj(space_path,testspace)

import pickle
from sklearn.naive_bayes import MultinomialNB  # 导入多项式贝叶斯算法包
def readbunchobj(path):
    file_obj = open(path, "rb")
    bunch = pickle.load(file_obj)
    file_obj.close()
    return bunch
# 导入训练集向量空间trainpath = "train_word_bag/tfidfspace.dat"train_set = readbunchobj(trainpath)
# d导入测试集向量空间testpath = "test_word_bag/testspace.dat"test_set = readbunchobj(testpath)
# 应用贝叶斯算法
# alpha:0.001 alpha 越小，迭代次数越多，精度越高clf = MultinomialNB(alpha=0.001).fit(train_set.tdm, train_set.label)
# 预测分类结果predicted = clf.predict(test_set.tdm)
total = len(predicted);rate = 0
for flabel, file_name, expct_cate in zip(test_set.label, test_set.filenames, predicted):
    if flabel != expct_cate:
        rate += 1
        print(file_name, ": 实际类别：", flabel, "-->预测分类：", expct_cate)
# 精度print("error_rate:", float(rate) * 100 / float(total), "%")

import numpy as np
from sklearn import metrics
#评估def metrics_result(actual,predict):
    print("精度：{0:.3f}".format(metrics.precision_score(actual,predict)))
    print("召回：{0:0.3f}".format(metrics.recall_score(actual,predict)))
    print("f1-score:{0:.3f}".format(metrics.f1_score(actual,predict)))
metrics_result(test_set.label,predicted)
中文文本语料
中文停用词文本集合
工程全部代码
原文链接