This is a data science project that displays the use of regression in machine learning. The project is based on a fictional real estate company who is seeking a data-driven approach to valuing properties.
The project is written in four different modules, each being a major step in the development of our machine learning solution.
Our client is a Real Estate Investment Trust (REIT) that is trying to predict the fair transaction price of a property before it's sold. The REIT's existing solution depends on third-party appraisal services, which is highly dependent on a professional's expertise. On average, inexperienced appraisers have been off by $70,000.
The REIT has a dataset of transaction prices for previous properties on the market. Our goal is to predict the transaction prices with an average error of $70k or less, which would then allow our client to replace inexperienced appraisers with our model.
Deliverable: Trained model file
Machine learning task: Regression
Target variable: Transaction Price
Win condition: Avg prediction error < $70k