Inspiration and Overview

Today, companies are using more and more of our data. Services like Facebook aren't actually free: we're paying with our data, and we don't know how it's used. We wanted to build an app that allows consumers to take control of their own data. Our app essentially creates your own profile of data by collecting data granted by the user. This data could be given to 3rd parties with the permission of the user. For example, if the user has a fitbit, they could grant access to the Fitbit api, and the app would produce advice and predictions for the user. This data will range from medical to financial, and in all these categories we would use this data to predict and give medical, financial, etc. advice to the user. Furthermore, none of the data would be sold to advertisers unless the user explicitly opts in, and the user would be paid a percentage of the money made.

What it does

We built a website that allows users to create an account and enter their data. So far, we have sample graphs for health data, and we provide feedback and recommendations based on the data. We also use machine learning to give predictions on health conditions such as diabetes and heart disease based on the data entered. We also used the Esri API so that we can get the information about the nearest coronavirus cases and tell the user how far the user is from the nearest virus. Watch that demo at

How we built it

For the front-end, we wrote in HTML from scratch, and use bootstrap to help us design the interface of the site. The graphs were produced using Plotly in Javascript.

For the back-end, we hosted the app using python and Flask, and we used flask-login and firebase to implement the account and login system. We hold every user in the firebase's Firestore, and hope to include user data there as well.

To implement the diabetes predictions, we used machine learning libraries such as scikit and pandas.

Challenges we ran into

We had some trouble creating the login system, since it was the first time we were doing it. We also struggled with sending data from Javascript to our Python backend.

Accomplishments that we're proud of

We produced a running website with a working login system that implements machine learning and can produce graphs.

What we learned

We learning a lot about time management, communication, and technical problem solving.

What's next for Delphi

We want to implement more applications of machine learning so our model can give predictions of specific features for the health, finance, etc.

Share this project: