You just got hired as a data scientist within a real estate property aggregation company that runs an aggregation website. Your organisation has a web app that aggregates properties for rent/sale from various agents across the country.
A week after your resumption, you get a mail from your boss with the following details:
Hello,
Welcome onboard, we need you to help us create a solution that addresses the issues customers have been having. Our customers are complaining about the user experience of the site. They have indicated that they need a rent budget estimator feature that gives them a quick estimate of their rent budget based on parameters such as location, size etc. which will be provided by them.
Based on the analysis recently done by the UI/UX team, the average customer typically needs to go through 10-20 web pages of properties to develop reasonable budget estimate. They believe this is not efficient and suggested we develop an algorithm that does this at the click of a button.
Also they indicated that we create another feature that recommends areas/houses to rent a house based on customer budget, size of house, desired commute time to work etc. The feature should be able to recommend top 5 options closest to the parameters customers provided.
- Plotly Dash for visualization and web apps.
- Python for machine learning.
- Using Python, you are expected to scrape the data from propertypro.ng.
- Develop a regression model that predicts house prices based on the data you scraped
- Using Plotly Dash, you are expected to create stunning interface which customers can interact with your models on.
Optional
- Create a recommender system that recommends areas/houses to rent a house based on customer budget, size of house and desired commute time to work
- ML model that predicts house prices
- Dash app that users can interact with to predict prices of house in the following areas Gbagada, Lekki Phase 1, Ajah, Ikorodu, Yaba, Surulere and Ikeja.