Inspiration
Helping first-time buyers from getting scammed
What it does
Predicts the price of a car depending on its VIN. It also shows a listing of vehicles based on preferences.
How we built it
Collected and cleaned over 10000 car listings to train our machine-learning model to give the most accurate responses possible. Hosted a website using React and Javascript to display our results
Challenges we ran into
Collecting the data was the biggest challenge. We wanted to get the highest accuracy, so we had to scrape many listings all over the internet.
Accomplishments that we're proud of
Achieved a low Mean Squared Error for our ML model, which consistently predicts the best price. We also have a user-friendly, reactive front-end UI
What we learned
We learned different types of regression models and what works best for various scenarios. We also learned how to set up a full-stack website.
What's next for AutoMatch
We plan to partner with car dealers to promote their listings but also ensure they provide our users with the best prices by using our model.
Log in or sign up for Devpost to join the conversation.