An LLM-based Intelligent System for the Evaluation of Property Geographical Environment

The real estate market is a complex system influenced by numerous factors, including geographical location, environmental amenities, and cultural preferences. This paper aims to investigate the role of environmental and geographical factors in real estate price prediction. We achieve this by leveraging the recent advances in multi-model large language model (MM-LLM), which has demonstrated remarkable capabilities in generating descriptive text from visual inputs. We developed a ranking system based on traditional Chinese cultural logic that assigns a score to each property based on its environmental attributes. Our findings indicate that properties with higher environmental rankings tend to have higher prices, particularly during market downturns.