/JamaicanHouse-Price-predictor

Price Prediction Model

Primary LanguageJupyter Notebook

Streamlit App

House-Price-prediction

Price Prediction Model

Intro:

This project is about predicting Jamaican house prices based on their features. My mother, a Reat Estate Agent, often gives her own price predictions of what a property might be worth and a good listing price based on given the market. I would like to develop a web app at first then eventually ship it to a mobile app, that way I can learn and practice more skills (for funsies).

Steps to completing this project:

  1. Gather property listings from Realtor Association of Jamaica's website. (Mainly houses and townhouses for now):

    • To get started, I scraped listings from RAJ's website. I am interested in the location, no. of bedrooms, no. of bathrooms and the sqft of each property. The listings were stored in a csv file and opened in Google Sheets. I did some initial data cleaning by encoding the regions based on their locations in Kingston & St.Andrew. Then I used Python to prep the data for modelling and produced some visualizations.

  2. Create a ML model to do the predicting

    • Next up, I experimented with ML algorithms to see which one best fit the data and generated the most accurate predictions. Again, this will initially only be for houses and townhouses in Kingston & St.Andrew.

  3. Learn a framework to create the web app

    • The app has been deployed using Streamlit! There's still work that needs to be done but I am so proud of what I have been able to achieve thus far!!!