SEMANTIC RETRIEVAL FOR RECOMMENDATION SYSTEMS WITH LARGE LANGUAGE MODEL
Abstract
Recommender systems (RS) are widely used to reduce information overload by predicting user preferences and suggesting relevant items based on historical data. Many traditional methods rely heavily on user-item interaction data, which can be sparse and difficult to reuse when deployed to different data domains. In this paper, we build a semantic recommendation pipeline based on text content vector representations. The system proposes a method for matching text features by domain, then uses a large language model (LLM) to create vector embedding to encode descriptive information of items combining text fields according to each data domain and retrieves recommendations based on cosine similarity to improve recommendation quality and efficiency. The pipeline is applied to four data domains: Books, Magazines, Movies, and Beauty, all of which support thumbnails for visualizing the results.
Tài liệu tham khảo
[1]. Sarwar, B., Karypis, G., Konstan, J., Riedl, J., (2001), “Item-based collaborative filtering recommendation algorithms”, Proceedings of the 10th International World Wide Web Conference (WWW), Trang 285 - 295.
[2]. Koren, Y., Bell, R., Volinsky, C., (2009), “Matrix factorization techniques for recommender systems”, IEEE Computer, Tập 42, Số 8, Trang 30 - 37.
[3]. Mnih, A., Salakhutdinov, R., (2008), “Probabilistic matrix factorization”, Advances in Neural Information Processing Systems (NeurIPS), Trang 1257 - 1264.
[4]. Funk, S., (2006), “Netflix update: Try this at home”, http://sifter.org/~simon/journal/20061211.html (truy cập ngày 19/04/2026).
[5]. Kim, Y., (2014), “Convolutional Neural Networks for Sentence Classification”, Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), Trang 1746 - 1751.
[6]. Vaswani, A., Shazeer, N., Parmar, N., et al., (2017), “Attention is All You Need”, Advances in Neural Information Processing Systems (NeurIPS), Trang 5998 - 6008.
[7]. Zhang, Y., et al., (2023), “Large Language Models for Recommender Systems: A Survey”, arXiv preprint arXiv:2308.10822.
[8]. Gao, J., et al., (2023), “A Survey on Large Language Models for Recommendation”, arXiv preprint arXiv:2305.19860.
[9]. Devlin, J., Chang, M.-W., Lee, K., Toutanova, K., (2019), “BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding”, Proceedings of NAACL-HLT, Trang 4171 - 4186.
[10]. Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., et al., (2020), “Language Models are Few-Shot Learners”, Advances in Neural Information Processing Systems (NeurIPS), Vol. 33, Trang 1877 - 1901.
[11]. Covington, P., Adams, J., Sargin, E., (2016), “Deep Neural Networks for YouTube Recommendations”, Proceedings of the 10th ACM Conference on Recommender Systems (RecSys), Trang 191 - 198.
[12]. Yue, Z., et al., (2023), “Large Language Models are Powerful Sequential Recommender Systems”, arXiv preprint arXiv:2308.11771.
[13]. L. Wang, N. Yang, X. Huang, L. Yang, R. Majumder, and F. Wei, “Multilingual E5 Text Embeddings: A Technical Report,” arXiv preprint arXiv:2402.05672, 2024.
[14]. Manning, C. D., Raghavan, P., Schütze, H., (2008), “Introduction to Information Retrieval”, Cambridge University Press.
[15]. Reimers, N., Gurevych, I., (2019), “Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks”, Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing (EMNLP), Trang 3982 - 3992.
[16]. Herlocker, J. L., Konstan, J. A., Terveen, L. G., Riedl, J. T., (2004), “Evaluating Collaborative Filtering Recommender Systems”, ACM Transactions on Information Systems, Tập 22, Số 1, Trang 5 - 53.