Final Result

Sustainability Carbon Footprint Estimator

About the Project

Our team created a sustainability-focused project for the macathon themed around Google's latest Artificial Intelligence and Machine Learning technologies. The goal was to build an application that helps users predict their carbon footprint for the month based on various personal habits and factors.

Inspiration

The inspiration behind this project stemmed from a growing interest in sustainability and the need to better understand our individual impact on the environment. With the increasing focus on climate change and the environment, we wanted to build a tool that would make it easier for people to estimate their carbon footprint and discover ways to reduce it.

How We Built the Project

We used Python to build the backend of our application, which calculates a person’s carbon footprint based on various inputs provided by the user. The inputs include:

Sex
Diet
Shower frequency
Transport type
Monthly grocery bill
Energy efficiency
etc. (in total 17 features)

We trained a machine learning model using Vertex AI and a dataset from Kaggle that estimates the carbon footprint for individuals in North America. After gathering the necessary user data, our model predicts the carbon footprint for the user over the course of a month.

Additionally, our app interfaces with Google's Gemini AI to provide personalized suggestions on how users can reduce their carbon footprint and live more sustainably. The recommendations from Gemini are then saved to a Google Doc for future reference.

Challenges Faced

The hardest part of the project was integrating multiple Google APIs, especially ensuring that each one functioned correctly in sync with the others. This involved:

Handling authentication and permissions for different Google APIs
Ensuring data flows smoothly between the application, Vertex AI, and Gemini
Debugging various issues related to API calls and data formatting

Despite these challenges, we were able to create a fully functional application that offers real-time predictions and actionable insights for users looking to reduce their environmental impact.

What We Learned

This project gave us valuable hands-on experience with Google's AI and ML technologies, particularly Vertex AI and Gemini. We learned how to:

Integrate machine learning models into a web-based application
Use APIs for real-time data analysis and communication
Overcome technical challenges related to data flow and API integration

Overall, the hackathon was a fantastic learning experience that enabled us to explore innovative solutions for promoting sustainability through technology.