Ethiclo

Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned## Inspiration

8-10% of global greenhouse gas emissions come from the fashion industry, selling between 80 and 150 billion items a year globally, and 60% of these items end up in incinerators or landfills. All of this results from the growth of 'Fast Fashion' when brands produce collections several times a year with no concern for the longevity of items or sustainable practices.

For context: Zara → produces 24 new collections per year H&M → produces 12-16 collections per year and refreshes them weekly Shein → updates collections and styles in real-time

As consumers, this is obviously frustrating. Even though 55% of consumers are interested in buying sustainable clothing, 48% don’t know where to find sustainable brands and alternatives.

What it does

Ethiclo is a tool designed to help you find sustainable alternatives during your online shopping journey.

Say you're browsing through a website for a black dress. You click on the best option at the lowest price, but see that it's made with unsustainable plastic-based materials. Ethiclo will then show you not only a 'sustainability score' based on various factors like materials and brand certifications, and show you more sustainable versions of that black dress on other websites.

From there you can save items into your own personal dashboard to reference later, and even check out the product directly by clicking on the link, making sustainable shopping that much easier.

How we built it

We used React to design the frontend as well as a combination of Flask, Postgres, and PyTorch for the backend elements including the item categorization and the sustainability score calculation.

Challenges we ran into

Given the size of our dataset to train the model to identify items of clothing, it took a long time to complete, and as with any machine learning, it was also a challenge to find enough of the right data that was relevant.

Accomplishments that we're proud of

Overall we're really happy with the way we were able to incorporate ML in two key aspects of our product, and more importantly, create something that customers would actually want to use day-to-day.