Alternate bike/transit route suggestions
Give individuals incentives, to support environmentally friendly initiatives.
What it does
Provides a user with a Green Score out of 100, and methods by which it can be improved. An insurance company can incentivize a customers premium based on the green score.
How we built it
We used Python and Scipy, to train and build a regression model using a users past data, such as accelerometer, proximity sensor, number of bike trips, number of public transport trips, car revving (yup!) and computed weights for a green score. (Since we did not have access to this data, we simulated it for testing, but built the pipeline). We exposed this data using a python flask api service and on the frontend, we created a react-app, to visualize the greenscore, suggest improvement techniques and provide alternate transportation options, when possible.
Challenges we ran into
Figuring out, what kind of data is feasible to collect in the real world, what kind of model to use and how to provide suggestions for green score improvement.
Accomplishments that we're proud of
A simple ui interface, that provides actionable items for a user to be more green!
What we learned
Some statistical packages in python, and minimalistic frontend design
What's next for gogreen
Integrate with real world data, from insurance companies, and test it on some opt-in users for a user-study.