Inspiration
Fast fashion has become a prevalent part of society, especially among teenagers who want the latest trends, that come and go in the blink of an eye, at a low price. Fast fashion is a term coined for the rapid production and consumption of clothing that has silently been wreaking havoc on our planet, exploiting laborers, and perpetuating a throwaway culture.
While brainstorming a solution, we understood that the main obstacle to consumers buying sustainable fashion was not a lack of awareness, but rather a lack of affordability and optimal user experience. So, we are proud to introduce... Conscious Closet!
What it does
After the user enters what item type, size, material, and color they want for their clothing, the website provides a link to a secondhand clothing item that matches what they want, along with statistics on the amount of water, electricity, and emissions they will save by buying secondhand. With minimal user input, our AI model is able to find an accurate, affordable, and sustainable alternative to clothing that would otherwise add more waste.
How we built it
We built the front-end of our website with HTML and CSS. Then, we added a Python back-end, in which we created a machine learning model to deploy on our website using the Flask API. This model is called a “K Nearest Neighbors” model, which uses the arguments the user enters to find the nearest sustainable item, providing the user with a link to it. We finished the website by adding hand drawn graphics to fit our theme and add a unique flair to our project.
Challenges we ran into
Artificial intelligence is tricky to make work, so we ran into problems where the dataset had missing data, or where the wrong value was targeted. However, we were able to solve this problem by using a default filler for the missing data, and changing the target value column.
Accomplishments that we're proud of
We're proud of being able to make the ML model work. It was a challenge, and required lots of effort, but it ended up working in the end. We're also proud of our website, made from scratch (including the graphics). We believe that it delivers the intended message, and is an easy way to access our ML model.
What we learned
We learned that there are many components that go into making a website, especially one that contains complicated back-end systems. We learned that there's more to hackathons than lines of code, and that design plays an especially important role.
What's next for Conscious Closet
Our team is planning on implementing image comparison to further improve the experience and accuracy. We hope to build a web scraper to add new clothes to our dataset. We hope to change our planet, one search at a time.
Log in or sign up for Devpost to join the conversation.