An Intelligent Tourism Recommender System Using Retrieval-Guided Large Language Models
Article Sidebar
Main Article Content
Abstract
Travel planning has become more complex nowadays due to the information overload from numerous online platforms causing travelers to be overwhelmed when selecting destinations, accommodations, and activities that match their preferences and constraints. Traditional tourism recommender systems usually provide static and generic suggestions that fail to match users’ needs and cannot understand natural user language. This research develops and evaluates a Generative AI Chatbot for Tourism Recommendations that use Large Language Models (LLMs) to generate personalized travel itineraries through natural language interaction to solve these limitations. The system can process user natural language inputs including travel origin and destination, trip duration and dates, and budget in a conversational way. This project used the osunlp/TravelPlanner dataset which contains real-world tourism information on attractions, restaurants, accommodations, and flights. The chatbot uses carefully designed prompts with strict generation rules to produce day-by-day itineraries that match user constraints. The preliminary results show the system’s capability to generate human–readable travel plans that include correct flight selections, accommodation recommendations, and structured daily activities. The chatbot prototype also supports conversational refinement which allows users to modify their itineraries interactively. The evaluation will be conducted using BLEU and NIST metrics to measure response quality while also using a user satisfaction survey to assess the usability and personalization of the itinerary in the next phase. This research has contributed to AI-driven tourism planning by showing how generative AI can be used to create interactive and context-aware assistants in recommender systems. Future work will focus on including external APIs and conducting a comprehensive user testing to measure the usability and satisfaction level in the real world.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
All articles published in JIWE are licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) License. Readers are allowed to
- Share — copy and redistribute the material in any medium or format under the following conditions:
- Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use;
- NonCommercial — You may not use the material for commercial purposes;
- NoDerivatives — If you remix, transform, or build upon the material, you may not distribute the modified material.
References
[1] S. Park and J. L. Nicolau, “Asymmetric effects of online consumer reviews,” Ann. Tour. Res., vol. 50, pp. 67–83, Jan. 2015, doi: 10.1016/j.annals.2014.10.007.
[2] F. Ricci, L. Rokach, and B. Shapira, “Recommender systems: Introduction and challenges,” in Recommender Systems Handbook, 2nd ed. Boston, MA, USA: Springer, 2015, pp. 1–34, doi: 10.1007/978-1-4899-7637-6_1.
[3] G. Adomavicius and A. Tuzhilin, “Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions,” IEEE Trans. Knowl. Data Eng., vol. 17, no. 6, pp. 734–749, Jun. 2005, doi: 10.1109/TKDE.2005.99.
[4] K.-Z. Liau and H. A. Santoso, “The role of generative AI in e-commerce recommender systems: Methods, trends and insights,” J. Informatics Web Eng., vol. 4, no. 3, pp. 35–63, Oct. 2025, doi: 10.33093/jiwe.2025.4.3.3.
[5] Z. B. Ter, P. Naveen, and J. Jayapradha, “Generative AI-based meal recommender system,” J. Informatics Web Eng., vol. 4, no. 2, pp. 315–338, Jun. 2025, doi: 10.33093/jiwe.2025.4.2.20.
[6] E. A. Anaam, S.-C. Haw, and P. Naveen, “Applied fuzzy and analytic hierarchy process in hybrid recommendation approaches for E-CRM,” JOIV: Int. J. Informatics Visualization, vol. 6, no. 2-2, pp. 553–560, Aug. 2022, doi: 10.30630/joiv.6.2-2.1043.
[7] D. Jannach, A. Manzoor, W. Cai, and L. Chen, “A survey on conversational recommender systems,” ACM Comput. Surv., vol. 54, no. 5, Art. no. 105, pp. 1–36, 2021, doi: 10.1145/3453154.
[8] A. Banerjee, A. Satish, F. N. Aisyah, W. Wörndl, and Y. Deldjoo, “Collab-REC: An LLM-based agentic framework for balancing recommendations in tourism,” ACM Trans. Recommender Syst., 2026, doi: 10.1145/3837862.
[9] W. X. Zhao et al., “A survey of large language models,” Front. Comput. Sci., vol. 20, Art. no. 2012627, 2026, doi: 10.1007/s11704-026-60308-3.
[10] M. L. Ferreira and E. Kastenholz, “Generative AI in hospitality and tourism: A dual-stakeholder perspective on tourist and workforce experience dynamics,” Tour. Rev., 2026, doi: 10.1108/TR-08-2025-0968.
[11] S. Mai and Z. Liu, “Generative AI and the tourist experience: Reconfiguring en-route and post-trip value co-creation and co-destruction,” Curr. Issues Tour., 2026, doi: 10.1080/13683500.2026.2640203.
[12] J. Wei et al., “Chain-of-thought prompting elicits reasoning in large language models,” in Advances in Neural Information Processing Systems, vol. 35, 2022, pp. 24824–24837.
[13] P. Lewis et al., “Retrieval-augmented generation for knowledge-intensive NLP tasks,” in Advances in Neural Information Processing Systems, vol. 33, 2020, pp. 9459–9474.
[14] X. Su and T. M. Khoshgoftaar, “A survey of collaborative filtering techniques,” Adv. Artif. Intell., vol. 2009, Art. no. 421425, 2009, doi: 10.1155/2009/421425.
[15] A. D. Solano-Barliza, I. Arregocés-Julio, M. Aarón-Gonzalvez, R. Zamora-Musa, E. De-La-Hoz-Franco, J. Escorcia-Gutierrez, and M. Acosta-Coll, “Recommender systems applied to the tourism industry: A literature review,” Cogent Bus. Manage., vol. 11, no. 1, Art. no. 2367088, 2024, doi: 10.1080/23311975.2024.2367088.
[16] V. Udandarao, N. A. Tiju, M. Vairamuthu, H. Mistry, and D. Kumar, “Roamify: Designing and evaluating an LLM based Google Chrome extension for personalised itinerary planning,” arXiv preprint arXiv:2504.10489, 2025, doi: 10.48550/arXiv.2504.10489.
[17] B. Liu, J. Ge, and J. Wang, “Vaiage: A multi-agent solution to personalized travel planning,” arXiv preprint arXiv:2505.10922, 2025, doi: 10.48550/arXiv.2505.10922.
[18] K. Otaki and Y. Baba, “Travel itinerary recommendation using interaction-based augmented data,” Expert Syst. Appl., vol. 269, Art. no. 126294, Apr. 2025, doi: 10.1016/j.eswa.2024.126294.
[19] A. Banerjee, A. Satish, F. N. Aisyah, W. Wörndl, and Y. Deldjoo, “SynthTRIPs: A knowledge-grounded framework for benchmark query generation for personalized tourism recommenders,” in Proc. 48th Int. ACM SIGIR Conf. Res. Develop. Inf. Retrieval (SIGIR ’25), Padua, Italy, 2025, pp. 3743–3752, doi: 10.1145/3726302.3730321.
[20] A. N. Fajari and A. Baizal, “Chatbot-based culinary tourism recommender system using named entity recognition,” JIPI (Jurnal Ilmiah Penelitian dan Pembelajaran Informatika), vol. 7, no. 4, pp. 1131–1138, 2022, doi: 10.29100/jipi.v7i4.3210.
[21] J. Xie, K. Zhang, J. Chen, T. Zhu, R. Lou, Y. Tian, Y. Xiao, and Y. Su, “TravelPlanner: A benchmark for real-world planning with language agents,” in Proc. 41st Int. Conf. Mach. Learn. (ICML), vol. 235, 2024, pp. 54590–54613.
[22] J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of deep bidirectional transformers for language understanding,” in Proc. 2019 Conf. North American Chapter Assoc. Comput. Linguistics: Human Language Technologies (NAACL-HLT), Minneapolis, MN, USA, 2019, pp. 4171–4186, doi: 10.18653/v1/N19-1423.
[23] T. B. Brown et al., “Language models are few-shot learners,” in Advances in Neural Information Processing Systems, vol. 33, 2020, pp. 1877–1901.
[24] A. Vaswani et al., “Attention is all you need,” in Advances in Neural Information Processing Systems, vol. 30, 2017, pp. 5998–6008.
[25] K. Papineni, S. Roukos, T. Ward, and W.-J. Zhu, “BLEU: A method for automatic evaluation of machine translation,” in Proc. 40th Annu. Meeting Assoc. Comput. Linguistics (ACL), Philadelphia, PA, USA, 2002, pp. 311–318, doi: 10.3115/1073083.1073135.
[26] E. Doostmohammadi, O. Holmström, and M. Kuhlmann, “How reliable are automatic evaluation methods for instruction-tuned LLMs?,” in Findings of the Association for Computational Linguistics: EMNLP 2024, Miami, FL, USA, 2024, pp. 6321–6336, doi: 10.18653/v1/2024.findings-emnlp.367.