What it does
Explainable AI that optimizes energy use in buildings.
How we built it
Using a SHAP-FOLD algorithm to run with xgboost with Android Studio as the frontend.
Challenges we ran into
The dependencies required to run swi-prolog and Java.
Accomplishments that we're proud of
Our frontend and our sleek algorithm.
What we learned
How to work with multiple normally-incompatible technologies.
What's next for Weather-based Building Power Management
We plan to add a web-based app running off of the expressjs backend so that clients can monitor and adjust their electricity consumption without having to download an app. We'd also like to improve our datasets and incorporate lighting and per-room temperature control into the algorithm.