AI Climate Trends and Solutions

Inspiration

Global warming is one of the biggest challenges that we face. Everybody agrees that we need to come together to ensure that future generations can continue living on Earth and use its resources. Our goal is to stop climate change, prevent impending doom, and thereby save humanity. But, baby steps. One thing that we have noticed, and you probably have too, is that people care less about issues if they can’t see the catastrophic effects. Sure, we all see reports of natural disasters on the news or hear about melting glaciers in the Arctic, but it is all too easy for people to turn off the TV or change the conversation without taking action to address the problem. We need to reduce our emissions. In order to do that, we need people to care.

What it does

This project serves as a connection between getting people to care and encouraging them to take action. By providing machine learning generated climate trends for cities around the world that can be accessed with an interactive map, we hope that users will see how global warming will affect them in the future and be inspired to implement solutions that will help us create a more sustainable reality. In order to support our users on their journey to saving the world, we made a quiz that will provide customized responses for sustainable practices they can implement, using a generative AI language model.

How we built it

All machine learning was completed using Python in a Google Colab (Jupyter Notebook). The implementation of the website, interactive map, and generative AI language model was completed using HTML, CSS, and JavaScript in Visual Studio Code.

Challenges we ran into

Machine Learning: We encountered a learning curve when it came to utilizing Python libraries to do what we wanted. Cleaning the data so that it was usable to train our model was the most time consuming part.
Interactive Map: We encountered a learning curve when it came to using a Scalable Vector Graphics file. We spent a lot of time thinking and discussing how to display the countries to improve user experience.
Generative AI Language Model: In order to engineer the prompt for the Generative AI Language Model, we had to discuss what and how to gather information for the data so we could generate personalized recommendations. Integrating the OpenAI API framework for ChatGPT was an extremely time consuming process, as we encountered many problems, like a paywall for use. However, in the end, we were able to successfully incorporate the Generative AI Language Model using JavaScript.

Accomplishments that we're proud of

We are very proud of being able to create an accurate prediction model for climate change and an interactive website that gives personalized solutions to create real change.

What we learned

In addition to the numerous technical skills we have gained (Machine Learning, Web Development, APIs), we have also learned about collaboration and teamwork. We all worked on different aspects of the project, but needed to communicate and work together to deliver the final product.

Built With

• ai
• colab
• css
• html
• javascript
• machine-learning
• openai
• python