Using Large Language Models (LLMs) for Data Collection and Standardization
Presentation Category
Modeling and Valuation
General Session Description
Data collection and standardization are often among the most time-consuming and costly parts of the assessment process. Better data lead to better valuations, improved public trust, and stronger defensibility of assessments. Early artificial intelligence in this field focused on deep learning models that extracted property characteristics from aerial and street-level imagery. This session will show how large language models such as ChatGPT and Claude can now be used within open-source software to assist with data collection and standardization from imagery. Examples include identifying features such as pools, property types, blight, depreciation, and solar panels. The goal is to demonstrate practical, low-cost methods that improve data quality and support fair, transparent, and efficient assessments.
Start Date
4-2-2026 9:00 AM
End Date
4-2-2026 10:30 AM
Using Large Language Models (LLMs) for Data Collection and Standardization
Data collection and standardization are often among the most time-consuming and costly parts of the assessment process. Better data lead to better valuations, improved public trust, and stronger defensibility of assessments. Early artificial intelligence in this field focused on deep learning models that extracted property characteristics from aerial and street-level imagery. This session will show how large language models such as ChatGPT and Claude can now be used within open-source software to assist with data collection and standardization from imagery. Examples include identifying features such as pools, property types, blight, depreciation, and solar panels. The goal is to demonstrate practical, low-cost methods that improve data quality and support fair, transparent, and efficient assessments.