vector_utils.py

# vector_utils.py

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity

def find_similar_docs(base_text, doc_dict, top_n=3):
    """
    base_text: 기준이 되는 요약 텍스트
    doc_dict: {filename: summary_text} 형태의 딕셔너리
    top_n: 추천 개수
    """
    docs = list(doc_dict.values())
    filenames = list(doc_dict.keys())

    texts = [base_text] + docs
    vectorizer = TfidfVectorizer().fit_transform(texts)
    vectors = vectorizer.toarray()

    similarities = cosine_similarity([vectors[0]], vectors[1:])[0]
    top_indices = similarities.argsort()[::-1][:top_n]

    return [(filenames[i], similarities[i]) for i in top_indices]