机器学习之新闻文本分类。

新闻文本分类首先需要通过大量的训练之后获得一个存放关键字的表，

之后再输入一个新闻内容，通过代码就可以自动判断出这个新闻的类别，

我这里是在已经有了新闻文本的关键词表后的处理，

# encoding=utf-8                                #遍历文件，用ProsessofWords处理文件
from imp import reload
import jieba
import os
import sys
from imp import reload
from sklearn.feature_extraction.text import CountVectorizer, TfidfTransformer
from sklearn.neighbors import KNeighborsClassifier


reload(sys)
VECTOR_DIR = 'vectors.bin'
MAX_SEQUENCE_LENGTH = 100
EMBEDDING_DIM = 200
TEST_SPLIT = 0.2


def deposit_txt(title, content):
    textpath = "news/news.txt"
    f = open(textpath, 'w+', encoding='utf-8')
    f.write(title+content)
    f.close()


def EnumPathFiles(path, callback, stop_words_list):
    if not os.path.isdir(path):
        print('Error:"', path, '" is not a directory or does not exist.')
        return
    list_dirs = os.walk(path)

    for root, dirs, files in list_dirs:
        for d in dirs:
            print(d)
            EnumPathFiles(os.path.join(root, d), callback, stop_words_list)
        for f in files:
            callback(root, f, stop_words_list)


def ProsessofWords(textpath, stop_words_list):
    f = open(textpath, 'r', encoding='utf-8')
    text = f.read()
    f.close()
    result = list()
    outstr = ''
    seg_list = jieba.cut(text, cut_all=False)
    for word in seg_list:
        if word not in stop_words_list:
            if word != '\t':
                outstr += word
                outstr += " "
    f = open(textpath, 'w+', encoding='utf-8')
    f.write(outstr)
    f.close()


def callback1(path, filename, stop_words_list):
    textpath = path + '\\' + filename
    print(textpath)
    ProsessofWords(textpath, stop_words_list)


def fenci():
    stopwords_file = "stopword/stopword.txt"
    stop_f = open(stopwords_file, "r", encoding='utf-8')
    stop_words = list()
    for line in stop_f.readlines():
        line = line.strip()
        if not len(line):
            continue
        stop_words.append(line)
    stop_f.close()
    print(len(stop_words))
    EnumPathFiles(r'news', callback1, stop_words)


def CV_Tfidf():

    reload(sys)

    # 数据获取
    print('(1) load texts...')
    train_texts = open('dataset_train/x_train.txt', encoding='utf-8').read().split('\n')
    train_labels = open('dataset_train/y_train.txt', encoding='utf-8').read().split('\n')
    test_texts = open('news/news.txt', encoding='utf-8').read().split('\n')
    all_text = train_texts + test_texts

    # 特征值抽取
    print('(2) doc to var...')

    count_v0 = CountVectorizer()
    counts_all = count_v0.fit_transform(all_text)
    count_v1 = CountVectorizer(vocabulary=count_v0.vocabulary_)
    counts_train = count_v1.fit_transform(train_texts)
    print("the shape of train is " + repr(counts_train.shape))
    count_v2 = CountVectorizer(vocabulary=count_v0.vocabulary_)
    counts_test = count_v2.fit_transform(test_texts)
    print("the shape of test is " + repr(counts_test.shape))

    tfidftransformer = TfidfTransformer()
    train_data = tfidftransformer.fit(counts_train).transform(counts_train)
    test_data = tfidftransformer.fit(counts_test).transform(counts_test)

    x_train = train_data
    y_train = train_labels
    x_test = test_data

    # KNN算法建模
    print('(3) KNN...')
    knnclf = KNeighborsClassifier(n_neighbors=3)
    knnclf.fit(x_train, y_train)
    preds = knnclf.predict(x_test)
    preds = preds.tolist()
    for i, pred in enumerate(preds):
        print(pred)
        if pred == '1':
            return"此新闻为娱乐类新闻"
        elif pred == '2':
            return "此新闻为汽车类新闻"
        elif pred == '3':
            return "此新闻为游戏类新闻"
        elif pred == '4':
            return "此新闻为科技类新闻"
        elif pred == '5':
            return "此新闻为综合体育最新类新闻"
        elif pred == '6':
            return "此新闻为财经类新闻"
        elif pred == '7':
            return "此新闻为房产类新闻"
        elif pred == '8':
            return "此新闻为教育类新闻"
        elif pred == '9':
            return "此新闻为军事类新闻"
def news(title, content):
    deposit_txt(title, content)
    fenci()
    result = CV_Tfidf()
    return result