Logo PTI Logo FedCSIS

Proceedings of the 21st Conference on Computer Science and Intelligence Systems (FedCSIS)

Annals of Computer Science and Information Systems, Volume 47

Enhancing Automated Valuation Models with LLM-based Feature Engineering and Spatial Indices and a RAG-Integrated Web Application for Trójmiasto Residential Market

,

DOI: http://dx.doi.org/10.15439/2026F2863

Citation: Piotr Kłopotowski,

Full text

Abstract. This paper presents a methodology for building robust Automated Valuation Models (AVMs) for residential properties in the Tricity (Gda\'nsk, Gdynia, Sopot) metropolitan area. Utilising a dataset of over 45,000 listings scraped over a nine-month period. We proposed an enriched feature engineering pipeline that integrates spatial indices (Points of Interest and public transport proximity) with unstructured data extracted via Large Language Models (LLMs). We evaluated the performance of XGBoost and Random Forest algorithms for both sale and rental price prediction. Results indicate that while traditional features like size and location remain dominant, the inclusion of custom spatial indices and LLM-extracted variables, such as utility fees for rentals, provides measurable improvements in model precision. Additionally, the paper focuses on a web application for analyzing residential market and outlines a practical application of RAG implementation for enhanced residential listing search based on the description of listings throught the web application.

References

  1. O. Aydin and K. Zieliński, Analysing the Residential Market Using Self-Organizing Map. Cham: Springer International Publishing, 2022, pp. 465–478. ISBN 978-3-030-85254-2. [Online]. Available: https://doi.org/10.1007/978-3-030-85254-2_28
  2. S. Rosen, “Hedonic prices and implicit markets: product differentiation in pure competition,” Journal of political economy, vol. 82, no. 1, pp. 34–55, 1974. https://dx.doi.org/10.1086/260169
  3. D. Tchuente, “Real estate automated valuation model with explainable artificial intelligence based on shapley values,” The Journal of Real Estate Finance and Economics, pp. 1–39, 2024. https://dx.doi.org/10.1007/s11146024-09998-9
  4. M. Fletcher, P. Gallimore, and J. Mangan, “Heteroscedasticity in hedonic house price models,” Journal of Property Research, vol. 17, no. 2, pp. 93–108, 2000. https://dx.doi.org/10.1080/095999100367930
  5. A. Valier, “Who performs better? avms vs hedonic models,” Journal of property investment & finance, vol. 38, no. 3, pp. 213–225, 2020. https://dx.doi.org/10.1108/jpif-12-2019-0157
  6. P. Kłopotowski, “Building prediction engine for residential property prices in tricity,” Master’s thesis, Gdańsk University of Technology, Gdańsk, Poland, 2026, field of study: Informatics (Machine learning). Supervisor: dr Olgun Aydin.
  7. H. Choi and H. Varian, “Predicting the present with google trends,” Economic record, vol. 88, pp. 2–9, 2012. https://dx.doi.org/10.1111/j.14754932.2012.00809.x
  8. O. Poursaeed, T. Matera, and S. Belongie, “Vision-based real estate price estimation,” Machine Vision and Applications, vol. 29, no. 4, pp. 667–676, 2018. https://dx.doi.org/10.1007/s00138-018-0922-2
  9. A. J. Pollestad, A. B. Næss, and A. Oust, “Towards a better uncertainty quantification in automated valuation models,” The Journal of Real Estate Finance and Economics, pp. 1–37, 2024. https://dx.doi.org/10.1007/s11146024-10002-7
  10. Y. Zhao, J. Zhao, and E. Y. Lam, “House price prediction: A multisource data fusion perspective,” Big Data Mining and Analytics, vol. 7, no. 3, pp. 603–620, 2024. https://dx.doi.org/10.26599/bdma.2024.9020019
  11. M. Steurer, R. J. Hill, and N. Pfeifer, “Metrics for evaluating the performance of machine learning based automated valuation models,” Journal of Property Research, vol. 38, no. 2, pp. 99–129, 2021. https://dx.doi.org/10.1080/09599916.2020.1858937
  12. P. Sobana, M. Balakumaran, S. Bharathkumar, P. Boopathi, and J. Harish, “House price prediction using machine learning,” in Challenges in Information, Communication and Computing Technology. CRC Press, 2024, pp. 704–708.
  13. J. Hong and W.-s. Kim, “Combination of machine learning-based automatic valuation models for residential properties in south korea,” International Journal of Strategic Property Management, vol. 26, no. 5, pp. 362–384, 2022. https://dx.doi.org/10.3846/ijspm.2022.17909
  14. Y. Jiang, L. Yue, S. Yang, and B. Yan, “Predicting real estate prices: A gradient boosting tree approach,” in Proceedings of the 2025 International Conference on Digital Transformation and Management, ser. DTM ’25. New York, NY, USA: Association for Computing Machinery, 2026. https://dx.doi.org/10.1145/3795546.3795587. ISBN 9798400721571 p. 288–293. [Online]. Available: https://doi.org/10. 1145/3795546.3795587
  15. M. Tekin and I. U. Sari, “Real estate market price prediction model of istanbul,” Real Estate Management and Valuation, vol. 30, no. 4, pp. 1–16, 2022. https://dx.doi.org/10.2478/remav-2022-0025
  16. R. Cellmer, M. Bełej, and R. Trojanek, “Housing prices and points of interest in three polish cities,” Journal of Housing and the Built Environment, vol. 39, no. 3, pp. 1509–1540, 2024. https://dx.doi.org/10.1007/s10901024-10124-7
  17. C. Chwiałkowski and A. Zydroń, “The impact of urban public transport on residential transaction prices: A case study of poznań, poland,” ISPRS International Journal of Geo-Information, vol. 11, no. 2, p. 74, 2022. https://dx.doi.org/10.3390/ijgi11020074
  18. N. Abhyankar, P. Shojaee, and C. K. Reddy, “Llm-fe: Automated feature engineering for tabular data with llms as evolutionary optimizers,” arXiv preprint https://arxiv.org/abs/2503.14434, 2025.
  19. S. R. Ahmad, “Enhancing multilingual information retrieval in mixed human resources environments: A rag model implementation for multicultural enterprise,” arXiv preprint arXiv:2401.01511, 2024.
  20. M. Arslan, H. Ghanem, S. Munawar, and C. Cruz, “A survey on rag with llms,” Procedia computer science, vol. 246, pp. 3781–3790, 2024. https://dx.doi.org/10.1016/j.procs.2024.09.178
  21. K. Ociepa, Łukasz Flis, K. Wróbel, A. Gwoździej, and R. Kinas, “Bielik 7b v0.1: A polish language model – development, insights, and evaluation,” 2024. [Online]. Available: https://arxiv.org/abs/2410.18565
  22. Minister Infrastruktury, “Rozporzadzenie ˛ ministra infrastruktury z dnia 12 kwietnia 2002 r. w sprawie warunków technicznych, jakim powinny odpowiadać budynki i ich usytuowanie,” Dz. U. 2002 nr 75 poz. 690, t.j. Dz. U. z 2022 r. poz. 1225, § 94, Poland, 2002.
  23. S. Es, J. James, L. E. Anke, and S. Schockaert, “Ragas: Automated evaluation of retrieval augmented generation,” in Proceedings of the 18th conference of the european chapter of the association for computational linguistics: system demonstrations, 2024, pp. 150–158.
  24. P. Rybak and M. Ogrodniczuk, “Silver retriever: advancing neural passage retrieval for polish question answering,” in Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), 2024. https://dx.doi.org/10.63317/364iqsayfxvs pp. 14 826–14 831.
  25. HuggingFace. sdadas/st-polish-paraphrase-from-mpnet. [Online]. Available: ttps://huggingface.com/sdadas/st-polish-paraphrase-from-mpnet
  26. J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “Bert: Pre-training of deep bidirectional transformers for language understanding,” in Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies, volume 1 (long and short papers), 2019, pp. 4171–4186.
  27. R. Kinas, P. Kiszczak, S. P. Perez, K. Ociepa, Ł. Flis, K. Wróbel, and A. Gwoździej, “Bielik-minitron-7b: Compressing large language models via structured pruning and knowledge distillation for the polish language,” arXiv preprint arXiv:2603.11881, 2026.