FlipEstate

Inspiration

We were inspired by the real estate market and how interest rates have increased. This led us to wonder how investing will work in new macro-economic conditions

What it does

It presents the users with properties on sale and the user can decide whether they want to save it for later or not. It puts a twist on the tedious experience of looking for properties. You never know what you will find next!

How we built it

We built it using Streamlit and databases to aid the linear regression ML model.

Challenges we ran into

We could not find extensive datasets that were anywhere comparable to leading apps. Moreover, it wasn't easy to use postcodes as a non-numeric category to indicate how prices will differ for similar houses across different postcodes.

Accomplishments that we're proud of

We are proud to have a functional web page that utilizes Google Maps API to show the street view for houses. Moreover, we are proud of the idea as it is less mundane than looking for houses on your own and surprises the user with unexpected results.

What we learned

We learned that it is severely difficult to process datasets for properties of all sizes since their location is vital to their prices but location properties are tricky to handle as values within the training dataset.

What's next for FlipEstate

FlipEstate will improve the recommendation API and make the predictions more robust and accurate. We will use user input to calculate the net profit margin from the Linear Regression ML model in real time. They will be able to change parameters in real time

Built With

• google-maps
• linear-regression
• machine-learning
• pandas
• python
• scikit-learn
• streamlit