We were inspired by the many arguments that alternative and renewable fuel sources are not 'sustainable', and how worrying it was for us that companies are choosing to pollute and continue to participate in environmental degradation, so we went out to prove that companies could have both profit and environmentally friendly energy sources.
What it does
The user can choose to simulate the energy usage of a make believe company over a set amount of years (to the user's discretion), and the machine learning algorithm will calculate the profit the company will make. It will show the user that renewable energies get cheaper, while non renewables will increase in cost due to their limited efficiency.
How I built it
We built it using a ]angular front end, and a node + express backend that executes python scripts, which then perform the machine learning algorithm. The python scripts use the scikit library.
Challenges I ran into
The biggest challenge we ran into was choosing which machine learning algorithm to implement. Many algorithms we could have used turned out to be either too complex (polynomial regression gave us weird numbers) or too simple.
Accomplishments that I'm proud of
Getting the graphs displayed was a challenge, but we eventually overcame it and were able to display the machine learning generated graphs.
What I learned
We learned how to accurately choose machine learning algorithms, however, more importantly we learned that for a large company, the choice between renewable energy and profit doesn't have to be one that must be made.
What's next for Sustainablr
An expansion in an Android app, or and iOS app.