Real Estate Valuation Assistant
Created by: Jeff Allen, Matt Sitto, & Mike Samson
Welcome to the demo of our 2026 Capstone project, where we trained ML models to be able to predict the value of single family home real estate properties using real transaction data across the United States — covering properties in over 40 U.S. states.
Use the Real Estate Valuation Assistant below to get a real-time valuation of a property in over 40 U.S. states.
View the Project Highlights section to see more details about the performance of our model across different states, as well as other important details.
Real Estate Value Predictor
Real Estate Valuation Assistant
Powered by ML · 40+ states supported
Hello! I'm an AI Real Estate Valuation Assistant. I can help you evaluate whether a home is fairly priced, overpriced, or undervalued — using machine learning models trained on real sales data across 40+ states. Are you looking at buying a home?
Press Enter to send · Shift+Enter for new line
Estimated Value:
Need more information to calculate
Property Location
Waiting for user input
Features Used for Inference
- Census Tract
- Waiting for user input
- Zip
- Waiting for user input
- Neighborhood Code
- Waiting for user input
- Year Built
- Waiting for user input
- Square Footage
- Waiting for user input
- Stories
- Waiting for user input
- Bathrooms
- Waiting for user input
- Garage
- Waiting for user input
- Pool
- Waiting for user input
- Assessment Total (yr 1)
- Waiting for user input
- Assessment Land (yr 1)
- Waiting for user input
- Assessment Impr (yr 1)
- Waiting for user input
- Assessment Total (yr 2)
- Waiting for user input
- Assessment Land (yr 2)
- Waiting for user input
- Assessment Impr (yr 2)
- Waiting for user input
Derived Features for Inference
- Assessment Drift
- Waiting for user input
- Drift Trend
- Waiting for user input
- Land Drift
- Waiting for user input
- Neighborhood Appreciation
- Waiting for user input