사용자 리뷰 감성분석 기반 하이브리드 영화 추천 시스템의 이론적 기반 및 실증적 검증 = Sentiment-Fused Hybrid Recommender System for Movie Domain: Integrating Aspect-Based Emotion Analysis with Collaborative-Content Filtering

다국어 초록 (Multilingual Abstract)

With the explosive growth of online content and products, recommender systems have become essential for helping users efficiently discover relevant items. While collaborative filtering has been the most widely used technique, relying solely on rating data often limits recommendation accuracy. This paper proposes a hybrid movie recommendation system that combines user ratings with sentiment scores extracted from review texts. Content-based filtering (CBF) uses TF-IDF vectors of genres and tags to calculate similarity between user profiles and movies, while item-based collaborative filtering (IBCF) predicts ratings by combining similar items’ scores with sentiment analysis results. Experiments show that the proposed hybrid model achieves superior recommendation performance compared to single models or models without sentiment integration.