According to the IPCC, the world will see catastrophic effects if temperatures climb to 2 degrees over current levels. This news is extremely distressing and the task of saving the earth is certainly not a small task. We feel that the only way to get there is if more individuals make a conscious effort to take on more eco-friendly choices.

What it does

Our web application allows users to search and view products that they want to buy, but what makes this different from any other e-commerce site is that we nudge consumer towards making a more environment-friendly choice. We rank products based on relevance but also on their eco-friendliness based on carbon emissions of the materials they're made of.

How we built it

We used the ebay API to retrieve relevant products and information on their materials. Given this information, we trained a machine learning classifier that gives a score that determines how eco-friendly a product is based on multiple factors such as material, title and description.

Challenges we ran into

Our biggest challenge was training the model given that there was a limited amount of training data available for eco-friendly products. We also had to do a lot of research on the carbon emissions of different materials and assess their eco-friendliness levels.

Accomplishments that we're proud of

This was a great learning experience for all of us, particularly in the field of environment science which was something we weren't previously familiar with. Also, we learned a lot from each other and probably laughed too much

What we learned

We learned how to deal with integrating different sources of information and finding meaningful insights. We also realized how large the effects are of even the smallest purchases!

What's next for eCO2Seek

We hope to train our model on a larger data set and see how well it can do with even more products

Share this project: