Skip to content

Utilizing machine learning, identify undervalued residential properties with high rental income potential. Optimize ROI and inform property acquisitions through feature-based analysis.

Notifications You must be signed in to change notification settings

casiasjc/Seattle_Real_Estate_Price_Prediction

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 

Repository files navigation

Analyzing King County’s Real Estate: Insights for REIT investors

Project Overview

This project aims to assist Silicon Sound Investments, an REIT specializing in technology-focused real estate, to identify and acquire promising properties in King County, Washington. The objective is to build a machine learning model that can predict the potential rental income and property value of residential buildings based on their location, size, age, and other relevant features.

Business Problem

The primary business problem that the project aims to solve is identifying undervalued residential properties in King County, Washington, that have potential for high rental income and future appreciation in value. The predictions generated by the machine learning model are important from a business perspective because they will allow Silicon Sound Investments to identify the most promising properties in King County, negotiate better deals with sellers, and maximize ROI for their stakeholders.

The Data

The project uses the King County House Sales dataset, which is stored in kc_house_data.csv in the data folder of the GitHub repository. The dataset consists of 21,613 observations and 21 variables, including information on the location, size, age, and sale price of residential properties in King County. A description of the column names can be found in column_names.md in the same folder.

Methodology

The project uses linear regression models and a random forest model to predict the rental income and property value of residential properties in King County. The performance of the models is evaluated using the R-squared metric.

Results

The highest R-squared score obtained from the linear regression models is 0.618, while the random forest model achieved an R-squared score of 0.745. The top predictors for the random forest model include grade, square footage of living area, latitude, waterfront, and longitude.

Conclusion

The machine learning models developed in this project can help Silicon Sound Investments to identify undervalued residential properties in King County with potential for high rental income and future appreciation in value. Regularly updating and iterating the model with new data and additional variables can bring a significant return on investment to the company.

About

Utilizing machine learning, identify undervalued residential properties with high rental income potential. Optimize ROI and inform property acquisitions through feature-based analysis.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published