GISValTech2025

Mass appraisal plays a crucial role in efficiently and equitably valuing large numbers of properties, but achieving fairness and accuracy remains a challenge. While traditional models like linear regression have been commonly used, machine learning has proven to be effective, offering several benefits and strengths. To better understand these methods, the IAAO AI Taskforce has been tasked with researching and comparing various AI methodologies. In this panel, we will take an international perspective, comparing datasets from multiple countries to explore how these AI models—such as Decision Trees, Random Forest, Gradient Boosting, and Neural Networks—can be adapted to different valuation environments. By examining case studies across diverse regions, we aim to highlight best practices in using AI to create accurate and equitable property values worldwide.

IAAO AI Task Force
Luc Hermans, Data Scientist, Netherlands Council for Real Estate Assessment*
Russ Thimgan, President, Thimgan & Associates, Inc.*
Daniel Fasteen, Ph.D., Senior Research Scientist, Tyler Technologies*
Luke Jorgensen, Data Scientist, Lee County Property Appraiser’s Office*
Joshua Jorgensen, Data Scientist, Lee County Property Appraiser’s Office*
Paul Bidanset, Ph.D., Founder, Center for Appraisal Research and Technology*
Joe Wehrli, Enterprise Architecture Manager, Tyler Technologies*
Michael McCord, Ph.D., Ulster University 
Pim Hessing, Data Analyst, Netherlands Council for Real Estate Assessment
John Valente, Appraisal Regional Manager, Tyler Technologies
Anita Ng, Assistant Commissioner (Corporate and Technology) Services, the Government of Hong Kong Special Administrative Region, China
 
*In attendance