Coherence Score验证LDA主题分类水平

from gensim.models import CoherenceModel
from gensim.corpora.dictionary import Dictionary
from gensim.models.ldamodel import LdaModel
from sklearn.feature_extraction.text import CountVectorizer
import pandas as pd
import nltk
from nltk.tokenize import word_tokenize
from multiprocessing import freeze_support

# 下载停用词和分词器所需的数据
#nltk.download('punkt')
nltk.download('stopwords')

def main():
    # 读取Excel文件
    data = pd.read_excel("Laos news overall.xlsx")

    # 假设您有一个名为 "cleaned_title" 的列，包含了新闻标题
    titles = data['内容'].tolist()

    # 初始化分词器和停用词
    stop_words = set(nltk.corpus.stopwords.words('english'))
    tokenizer = nltk.RegexpTokenizer(r'\w+')

    # 分词并去除停用词
    tokenized_titles = []
    for title in titles:
        words = tokenizer.tokenize(title)
        words = [word for word in words if word.lower() not in stop_words]
        tokenized_titles.append(words)

    # 创建字典
    dictionary = Dictionary(tokenized_titles)
    corpus = [dictionary.doc2bow(title) for title in tokenized_titles]

    # 训练 LDA 模型
    num_topics = 10
    lda = LdaModel(corpus, num_topics=num_topics, id2word=dictionary)

    # 计算 coherence score
    coherence_model = CoherenceModel(model=lda, texts=tokenized_titles, dictionary=dictionary, coherence='c_v')
    coherence_score = coherence_model.get_coherence()

    print(f"Coherence Score: {coherence_score}")

if __name__ == '__main__':
    freeze_support()
    main()