EXPRESS: An LLM-Generated Explainable Property Valuation System with Neighbor Imputation

2025-03-16

AI Summary

To address two practical bottlenecks in real estate valuation—poor robustness to missing data and insufficient model interpretability—this paper proposes an interpretable valuation framework integrating dynamic nearest-neighbor imputation with large language model (LLM)-driven feature attribution. We innovatively design a rural–urban adaptive dynamic nearest-neighbor search mechanism, enabling user-configurable, fine-grained missing-value imputation. Furthermore, we leverage LLMs to generate natural-language explanations of feature importance, emulating professional human appraisal reasoning. Empirical evaluation on real-world property datasets demonstrates high prediction accuracy, strong robustness across diverse missing-data patterns, and compliance with financial regulatory requirements. To the best of our knowledge, this is the first work to systematically integrate dynamic nearest-neighbor retrieval with LLM-based attribution for real estate valuation—achieving a balanced trade-off among practicality, transparency, and regulatory compliance.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

The demand for property valuation has attracted significant attention from sellers, buyers, and customers applying for loans. Reviews of existing approaches have revealed shortcomings in terms of not being able to handle missing value situations, as well as lacking interpretability, which means they cannot be used in real-world applications. To address these challenges, we propose an LLM-Generated EXplainable PRopErty valuation SyStem with neighbor imputation called EXPRESS, which provides the customizable missing value imputation technique, and addresses the opaqueness of prediction by providing the feature-wise explanation generated by LLM. The dynamic nearest neighbor search finds similar properties depending on different application scenarios by property configuration set by users (e.g., house age as criteria for the house in rural areas, and locations for buildings in urban areas). Motivated by the human appraisal procedure, we generate feature-wise explanations to provide users with a more intuitive understanding of the prediction results.

Problem

• Handles missing data in property valuation systems
• Improves interpretability of property valuation predictions
• Customizes neighbor imputation based on user-defined criteria

Innovation

• LLM-generated explainable property valuation system
• Customizable missing value imputation technique
• Dynamic nearest neighbor search for similar properties