리뷰 데이터를 활용한 POI 추천 시스템
POI Recommender System Using Review Data
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
최근 자연어 처리 기술과 대형 언어 모델의 발전은 사용자 리뷰와 같은 텍스트 데이터를 활용한 연구 발전 속도를 촉진시키고 있다. 본 논문에서는사용자 리뷰 데이터를 결합하여 추천 성능을 향상시키는 POI(Point of Interest) 추천 시스템을 제안한다. 기존의 협업 필터링(CollaborativeFiltering) 모델은 주로 사용자-아이템 상호작용 데이터를 기반으로 하여 추천을 수행하지만, 이러한 방식은 데이터 희소성 문제로 인해 사용자의세부적인 선호도를 포착하는 데 한계가 있었다. 이를 해결하기 위해 본 연구에서는 사용자 리뷰 데이터를 대형 언어 모델을 활용하여 임베딩으로변환하고, 이를 협업 필터링 모델에 결합하여 선호도를 예측하였다. 실험 결과, Yelp 데이터셋에서 텍스트 데이터를 결합한 모델들이 HitRatio@K,Recall@K, Precision@K와 같은 주요 평가 지표에서 기존 모델보다 우수한 성능을 보였다. 또한, 추천된 POI와 관련된 구체적인 리뷰를 함께 제공함으로써 사용자에게 추천 결과의 신뢰성을 높이는 방법을 제시한다.
Abstract
With recent advances in natural language processing (NLP) and the development of large language models (LLMs), leveraging textualdata such as user reviews has significantly improved the performance of recommender systems. In this paper, we propose a Point ofInterest (POI) recommender system that enhances recommendation accuracy by integrating user review data. We employ a fine-tunedBERT (Bidirectional Encoder Representations from Transformers) model to extract embeddings from user reviews and integrate them withcollaborative filtering models (BPR-MF, LightGCN) to improve performance. Experimental results on the Yelp dataset show that modelsincorporating textual data outperform baseline models in key metrics such as HitRatio@K, Recall@K, and Precision@K. Additionally, byproviding users with relevant reviews alongside recommendations, the system enhances the transparency and trustworthiness of the results. This research demonstrates the potential of utilizing textual data in improving POI recommendation systems and offers new opportunitiesto enhance recommendation accuracy and user experience.
참고문헌 (21)
- R. He, "VBPR: visual bayesian personalized ranking from implicit feedback", Proceedings of the AAAI Conference on Artificial Intelligence, 30(1) (2016)
- L. Chen, "Revisiting graph based collaborative filtering : A linear residual graph convolutional network approach", Proceedings of the AAAI Conference on Artificial Intelligence, 34(1), pp.27~34 (2020)
- X. He, "Lightgcn: Simplifying and powering graph convolution network for recommendation", Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, pp.639~648 (2020)
- X. Wang, "Neural graph collaborative filtering", Proceedings of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval, pp.165~174 (2019)
- M. Zhou, "Description generation for points of interest", 2021IEEE 37th International Conference on Data Engineering (ICDE), pp.2213~2218 (2021), IEEE
- R. Baral, "ReEL", Proceedings of the 26th Conference on User Modeling, Adaptation and Personalization (2018), ACM
- X. He, "Trirank : Review-aware explainable recommendation by modeling aspects", Proceedings of the 24th ACM International on Conference on Information and Knowledge Management, pp.1661~1670 (2015)
- Y. Cui, "ASKAT : Aspect Sentiment Knowledge Graph Attention Network for Recommendation", Electronics, 13(1), pp.216~ (2024)
- G. Li, "Group-based recurrent neural networks for POI recommendation", ACM Transactions on Data Science, 1(1), pp.1~18 (2020)
- Z. Qiu, "U-BERT: Pre-training user representations for improved recommendation", Proceedings of the AAAI Conference on Artificial Intelligence, 35(5), pp.4320~4327 (2021)
- Y. Yang, "Lightweight composite re-ranking for efficient keyword search with BERT", Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining, pp.1234~1244 (2022)
- G. Penha, "What does bert know about books, movies and music? probing bert for conversational recommendation", Proceedings of the 14th ACM Conference on Recommender Systems, pp.388~397 (2020)
- L. Wu, "A survey on large language models for recommendation", World Wide Web, 27(5), pp.60~ (2024)
- R. de Brito Duarte, "AI Trust: Can Explainable AI Enhance Warranted Trust?", Human Behavior and Emerging Technologies, 2023(1), pp.4637678~ (2023)
- X. Wang, "A reinforcement learning framework for explainable recommendation", 2018 IEEE International Conference on Data Mining (ICDM), pp.587~596 (2018), IEEE
- Y. Li, "Fairness in recommendation : Foundations, methods, and applications", ACM Transactions on Intelligent Systems and Technology, 14(5), pp.1~48 (2023)