Toyota Analyza | Devpost

Presenting, Toyota Analyza.
Our design, taking careful note of Toyota's design guide and official colors.
Our extensive filtering options
Our Google Colab where we developed our models in Python

Inspiration

Getting to talk to Toyota about the challenge and getting to see the high level of expertise they carry over these fields inspired us to take on this challenge. What kept us inspired are the discoveries we made along the way during the analysis.

What it does

Complex analysis on 5 years worth of data on fuel economy displayed for employees to view easily. At the touch of a button, apply filters to narrow down (or widen!) your search.

Want to see the PPM of gas vehicles from 2021 to 2023? We got you. What if you want to include hybrid vehicles as well? With a tap, its done.

Our project showcases feature analysis on more than 11k lines of data to gain insights on the most important trends using machine learning and our in-house API calls.

How we built it

At first, we were unsure on how to approach the task at hand. After brainstorming amongst ourselves, we made our first visit to the Toyota booth, where everyone was extremely helpful in clarifying and answering our questions. With new inspiration and feedback in mind, we planned and designed our solution.

We designed the mock-up of our project in Figma, and coded the front-end fully in HTML, CSS, and Javascript. On the back-end side of things, we put combined the datasets with MySQL and utilized Python with scikit-learn to develop two algorithms, a LLM, and a custom API to retrieve data from the dataset. Our in-house API is built with Flask and tested with Postman. Once time came to merge the two together, we maneuvered to Tailwind and Vue for easier communication between both ends.