Inspiration
Going into the hackathon, we wanted to work on a project that truly will push us to our limits in terms of skill and creativity. We stumbled upon the EOG challenge, which we found quite intriguing.
What it does
- Optimizes the witch collection schedule per day by delivering the daily potion production to the market.
- Uses machine learning with scikit-learn to predict the probability of a collection event occurring.
- Utilizes logistic regression machine learning to predict the probability of an overflow occurring.
How we built it
- React for the frontend
- Python for the backend
- Matplotlib for visualizing the data
Challenges we ran into
- Decoding the challenge and understanding the core ideas of the prompt was only half the battle. Figuring out the logic for the route optimization was rather difficult at first, as we had to use our prior knowledge and construct a new algorithm from that. We were able to solve this challenge through analyzing the data through insights and understanding the patterns within the data.
Accomplishments that we're proud of
- Creating a frontend that is on-theme with both the hackathon and the problem statement.
- Generating an algorithm that is able to efficiently optimize the routes taken.
- Making a fluid and dynamic UI.
What we learned
- Working with a project which requires a deep understanding of the subject matter.
- Applying mathematical and statistics concepts in order to achieve the main goal.
What's next for WitchWatch
- Adding more features, such as an optimization of the algorithms implemented and a text message notification system that sends a summary of the data to the user at the end of the day.
Log in or sign up for Devpost to join the conversation.