中文分词（上）——获取和Word2Vec模型构建

首先，我们先完成读取一篇文本文档，去除stopwords，只保留中文字符后进行分词。以下代码包括两个函数：

建议在看此部分代码之前，先看看数据集的层次结构，方便代码理解

import re,os,jieba
import numpy as np
import pandas as pd
import jieba.analyse

def extract_words_one_file(filepath,all=True):

    def open_file(file_text):
        with open(file_text,'r',errors='ignore',encoding='utf-8') as fp:
            content=fp.readlines()
        return content

    def remove(text):
        remove_chars=r'^\u4e00-\u9fa5'
        return re.sub(remove_chars,'',text)

    def open_stop(file_stop):
        stopwords=[line.strip() for line in open(file_stop,'r',encoding='utf-8-sig').readlines()]
        return stopwords

    def seg_sentence(sentence):
        sentence_seged=jieba.cut(sentence.strip())
        stopwords=open_stop('data/stopwords/hit_stopwords.txt')
        outstr=''
        for word in sentence_seged:
            if word not in stopwords:
                outstr+=word
                outstr+=' '

        return outstr.strip()

    inputs=open_file(filepath)

    words_in_sentence=[]
    for line in inputs:
        line_delete=remove(line)

        line_seg=seg_sentence(line_delete)
        words_in_sentence.append(line_seg)

    print('words_in_sentence_1:',words_in_sentence)
    words_in_sentence=[x for x in words_in_sentence if x!='']
    print('words_in_sentence_2:', words_in_sentence)

    alltokens=[]
    chinesewords_sentence=[]
    for i in range(len(words_in_sentence)):

        word=re.split(r'\s',words_in_sentence[i])
        alltokens.append(word)

    print('alltokens:',alltokens)

    for element in alltokens:
        element=[x for x in element if x!='']
        chinesewords_sentence.append(element)

    print('chinesewords_sentence:',chinesewords_sentence)

    chinesewords_article=[i for k in chinesewords_sentence for i in k]

    print('chinesewords_article:',chinesewords_article)

    if all==True:
        return chinesewords_article
    else:
        return chinesewords_sentence

def extract_words_folder(path, all=True):

    files=os.listdir(path)
    features=[]

    for i in range(len(files)):
        dirs=os.listdir(path+'/'+files[i])

        for f in dirs:
            if all==True:
                word_single_text=extract_words_one_file(path+'/'+files[i]+'/'+f,all=True)
                word_with_label = [word_single_text, files[i], f]
                features.append(word_with_label)
            else:
                word_single_text = extract_words_one_file(path + "/" + files[i] + "/" + f, all=False)
                features.append(word_single_text)

    if all == True:
        return pd.DataFrame(features, columns=['Words', 'Category', 'File'])
    else:
        return features

article_features = extract_words_folder(path='data/fudan-utf8/train',all = True)

article_features.to_csv("article_features_train_raw.csv",encoding='utf_8',index=False)

sent_features = extract_words_folder(path='data/fudan-utf8/train',all = False)
with open("word_sentence_train.txt", "w", encoding='utf-8') as f:
    f.write(str(sent_features))

为了方便之后数据的一致性，我没有将所有的数据都一并保存，之后再划分训练集测试集，而是之前手动划分，自己创建了一个test的文件夹，分别提取训练集和测试集的数据，然后分别存入了两个csv文件当中。我在划分的时候也发现，不同类别的样本数量差别很大，有的上千，有的还不过百。因此，我只选取了其中最多的九个类别，去除了剩余的11个类别。因为这些样本数量过少的类别相比于其他悬殊的样本数量，很难让模型作出无偏见的判定。

import pandas as pd
import numpy as np

train=pd.read_csv('data/article_features_train_raw.csv')
test=pd.read_csv('data/article_features_test_raw.csv')

train.Category.replace('C31-Enviornment','C31-Environment',inplace=True)
train=train[(train['Category'] == 'C3-Art')|(train['Category'] == 'C11-Space')|(train['Category'] == 'C19-Computer')
             |(train['Category'] == 'C31-Environment')|(train['Category'] == 'C32-Agriculture')
             |(train['Category'] == 'C34-Economy')|(train['Category'] == 'C38-Politics')|(train['Category'] == 'C39-Sports')
             |(train['Category'] == 'C7-History')]
test = test[(test['Category'] == 'C3-Art')|(test['Category'] == 'C11-Space')|(test['Category'] == 'C19-Computer')
             |(test['Category'] =='C31-Environment')|(test['Category'] == 'C32-Agriculture')
             |(test['Category'] == 'C34-Economy')|(test['Category'] == 'C38-Politics')|(test['Category'] == 'C39-Sports')
             |(test['Category'] == 'C7-History')]

label2category = {0: 'C11-Space', 1: 'C19-Computer', 2: 'C3-Art', 3: 'C31-Environment', 4: 'C32-Agriculture',
                  5: 'C34-Economy', 6:'C38-Politics',7:'C39-Sports',8:'C7-History'}
category2label = dict(zip(label2category.values(), label2category.keys()))

train['label'] = train.Category.replace(category2label)
test['label'] = test.Category.replace(category2label)
train = train.reset_index(drop=True)
test = test.reset_index(drop=True)

train.to_csv("article_features_train.csv",encoding='utf_8_sig',index=False)
test.to_csv("article_features_test.csv",encoding='utf_8_sig',index=False)

分词提取完毕后，我们需要用一定的方式表示这些分词，将这些计算机无法处理的非结构化信息转化为可计算的结构化信息。one-hot方法是其中之一，它的原理很好理解，首先生成一个初始值为零的长度为所有词的列表。若文章或句子包含某词，则该词对应位置为1，否则为0。但是其在表示大规模文本的时候往往出现过于稀疏的现象，也无法表示词语与词语间的关系，所以只是在比较简单的实践中使用。

本文我们所采取的文本表示方法是词嵌入中的Word2Vec。通俗的来说，就是根据训练将分词用多维向量表示。其2种训练模式为通过上下文来预测当前词和通过当前词来预测上下文。

gensim版本为4.1.2

import gensim

with open("word_sentence.txt", "r") as f:
    word_sentence = f.read()

sent_feature = eval(word_sentence)

sent_words = [i for k in sent_feature for i in k if len(i)>3]

在第二部分中，我们已经将分词按句子形式保存到了文本文件中，现在，我们便利用它直接进行训练获得模型。

model = gensim.models.Word2Vec(sent_words, sg=1, size=100, window=3,iter=5,
                               min_count=3, negative=3, sample=0.001, hs=1)

model.wv.save_word2vec_format('./word2vec_model.txt', binary=False)

w2v_model=gensim.models.Word2Vec(vector_size=300,window=3,sg=1,min_count=3)
w2v_model.build_vocab(sent_words)

third_model=gensim.models.KeyedVectors.load_word2vec_format('data/fudan-utf8/sgns.merge.word',binary=False)

w2v_model.build_vocab([list(third_model.vocab.key_to_index())],update=True)
w2v_model.intersect_word2vec_format('data/fudan-utf8/sgns.merge.word',binary=False,lockf=1.0)
w2v_model.train(sent_words, total_examples=w2v_model.corpus_count, epochs=5)
print("Model training finished.")

w2v_model.wv.save_word2vec_format('./word2vec_ensemble.txt', binary=False)
print("Model saved.")

Original: https://blog.csdn.net/qq_42801194/article/details/121535761
Author: Vincy_King
Title: 中文分词（上）——获取和Word2Vec模型构建