Using Regional and Socioeconomic Features to Identify Unbalanced Home Pricing
We leveraged commonly unused attributes to identify previously unseen trends in housing prices across the country including weather, crime rates, proximity to schools, etc outside of just median home value in that Zipcode.
Machine Learning
Python
Project Overview
Project Name: US Real Estate Pricing Model
Description: For my Data Mining coursework, I developed a machine learning model designed to expose unbalanced pricing within the highly volatile US real estate market. By scraping and aggregating property data into a PostgreSQL database, I built a predictive pipeline in Python that flags localized market inefficiencies and pricing anomalies. The project ultimately served as a data-driven tool to help identify properties that deviate significantly from expected market values, translating complex data mining techniques into practical real estate insights.
Technologies: Python, PostgreSQL
Timeline: Spring 2026
Role: Lead Architect
How to Access:
Slides:https://docs.google.com/presentation/d/1o7Tgbza7LAqajRISrmPG30B7kIjRUCX5GEmx50pvI_I/edit?usp=sharing
Report: https://drive.google.com/file/d/1xfwOlC_FaC3CQoUdVVK8qfYvOJuksk9b/view?usp=sharing
Github: https://github.com/kkim-4/us-zipcode-fair-pricing-model