NLP 用 Python を使用して、複数の PDF ファイルから類似のテキストをすばやく抽出するにはどうすればよいですか?

pip install PyPDF2 nltk gensim

import PyPDF2

def extract_text_from_pdf(file_path):
    with open(file_path, 'rb') as file:
        reader = PyPDF2.PdfFileReader(file)
        text = []
        for page_num in range(reader.numPages):
            page = reader.getPage(page_num)
            text.append(page.extract_text())
        return ' '.join(text)

# 示例用法
file_path = 'path/to/pdf/file.pdf'
text = extract_text_from_pdf(file_path)
print(text)

import nltk
from nltk.corpus import stopwords
from nltk.tokenize import word_tokenize
from nltk.stem import WordNetLemmatizer
import string

def preprocess_text(text):
    # 分词
    tokens = word_tokenize(text)
    
    # 转换为小写字母
    tokens = [token.lower() for token in tokens]
    
    # 去除停用词
    stop_words = set(stopwords.words('english'))
    tokens = [token for token in tokens if token not in stop_words]

    # 去除标点符号和数字
    tokens = [token for token in tokens if token not in string.punctuation and not token.isdigit()]

    # 词形还原
    lemmatizer = WordNetLemmatizer()
    tokens = [lemmatizer.lemmatize(token) for token in tokens]
    
    # 合并词汇
    text = ' '.join(tokens)
    
    return text

# 示例用法
preprocessed_text = preprocess_text(text)
print(preprocessed_text)

from gensim import corpora, models, similarities

def compute_similarity(texts):
    # 创建词袋模型
    dictionary = corpora.Dictionary(texts)
    corpus = [dictionary.doc2bow(text) for text in texts]
    
    # 计算TF-IDF
    tfidf = models.TfidfModel(corpus)
    tfidf_corpus = tfidf[corpus]
    
    # 计算相似度矩阵
    index = similarities.MatrixSimilarity(tfidf_corpus)
    
    # 计算相似文本
    similarities = index[tfidf_corpus]
    
    return similarities

# 示例用法
texts = [preprocess_text(text1), preprocess_text(text2), preprocess_text(text3)]
similarity_matrix = compute_similarity(texts)
print(similarity_matrix)

def find_similar_texts(texts, threshold):
    similar_texts = []
    for i in range(len(texts)):
        for j in range(i+1, len(texts)):
            if similarity_matrix[i][j] > threshold:
                similar_texts.append((i, j))
    return similar_texts

# 示例用法
similar_texts = find_similar_texts(texts, 0.7)
for i, j in similar_texts:
    print(f'Text {i+1} is similar to Text {j+1}')