Inspiration
Bayer LifeHubs challenge and their interest in an intelligent solution to calculating potential gains from reducing carbon emissions drove us to pursue our project, Carb・AI
What it does
Carb・AI aims to aggregate emission data from companies in addition to generating its own evaluation of company emissions, in order to bring greater awareness to both consumers and businesses on the carbon emission landscape (i.e. which companies have taken not only actionable but resultant steps to achieve lower carbon emissions).
How we built it
Using Djano & Bootstrap for our website, and JS for our presentation.
Challenges we ran into
We had planned on performing data analysis using pandas/tensorflow but due to being unable to find any reasonable, usable data, pivoted to our business model of being the ones that would create the data.
Accomplishments that we're proud of
We are proud of identifying through our research the lack of transparent, readily available data for carbon-conscious individuals and firms, and shedding light on the issues both economical and social that this causes.
What we learned
Although we invested many hours in trying to find usable carbon emissions data to create models that would help us determine a companies carbon footprint, we realize that the problem lay within getting such data in the first place. Especially as it relates to data science, we realized the importance of concretely defining a research question, and being sure the data or feature to answer such a question exists or can be engineered. Good data is hard to come by and is an important step in the machine learning process that is often overlooked.
What's next for Carb・AI
To continue searching for data to provide decent carbon footprint estimates, and looking for ways in which we can collect such data, either with or without industries own abilities, to provide meaningful and transformative data.
Log in or sign up for Devpost to join the conversation.