AI-Enhanced Housing Appraisal: Using Machine Learning to Quantify Curb Appeal in Real Estate Valuation
Thomas Youle
Faculty Mentor
Thomas Youle (Kelley School of Business)
Project Description
Can artificial intelligence help us better predict house prices by "seeing" what makes a home appealing? Traditional methods for estimating home values rely on basic features like square footage, number of bedrooms, and location. However, these miss important visual factors that influence what buyers are willing to pay—things like curb appeal, architectural style, and overall attractiveness. This project will test whether AI image analysis can capture these "hard-to-measure" qualities and improve housing price predictions. You'll work with real housing data from the Bloomington area, use AI tools to analyze property photos, and build statistical models that combine traditional features with AI-generated visual assessments.
Technology or Computational Component
AI Image Analysis: Using generative AI platforms to systematically evaluate property photographs and generate standardized scores for subjective characteristics like curb appeal and architectural quality Statistical Software: Implementation of hedonic regression models to compare traditional valuation methods with AI-enhanced approaches Data Management: Working with housing datasets, image databases, and AI service integrations.