Skip to content

While the assessed value is a strong predictor of a property's sales price, other variables will be a factor. Our tool takes an assessed value, acreage, class, and vacancy to predict sale price.

Notifications You must be signed in to change notification settings

MaxBrowning/butler-project3-groupD

Repository files navigation

Property Value Prediction With Machine Learning

Image of Project Home Page

Collaborators:

Project Overview:

Using collected data from Indiana Property Sales in 2020, machine learning was used to create a model that would allow property value prediction based on certain parameters such as assessed value, plot size and class of property. This information was collected via user input and calculated what the estimated price would be. Along side of the prediction tool is a dashboard that allows the user to interact and explore the collected data. Additionally, a comparison dashboard allows the user to select actual properties by certain filters and see the actual price of property sales from Indiana 2020 sales.

Dataset:

Indiana 2020 Property Sales Disclosure Form Data

Representative Images of Deployed Project:

Property Value Assessment Form:

Image of Property Value Assessment Form

Tableau County Dashboard:

Image of Tableau dashboard exploring county data

Feature Matrix Heatmap:

Image of Machine Learning feature matrix heatmap


Languages and Tools:

  • Python
  • HTML
  • CSS
  • SQL
  • Flask API
  • SKLearn
  • Pickle
  • Microsoft Excel
  • SQLAlchemy
  • Pandas
  • Jupyter Notebook
  • Matplotlib
  • Heroku

About

While the assessed value is a strong predictor of a property's sales price, other variables will be a factor. Our tool takes an assessed value, acreage, class, and vacancy to predict sale price.

Topics

Resources

Stars

Watchers

Forks