Chevron 2023 Datathon

Inspiration

As members of Duncan College and Rice University, eco-friendliness is big part of our community. With our composting bins and gardens, we were inspired by Chevron's interest in clean and renewable energy.

What it does

We created a machine learning model that predicts the amount of money a U.S. state will invest in renewable energy for a given year. We also analyzed the results across the country and came to conclusions about which states were the most anticipated to be the most eco-friendly.

How we built it

We cleaned and preprocessed data in Python and Pandas, and we created a random forest regression model in scikit-learn. We also created plots using plotly and matplotlib.

Challenges we ran into

Using Google colab was difficult because it did not update live between members, causing some rollbacks in the code. Additionally, we were exposed to new technologies; for example, collaboration software and plotting frameworks.

Accomplishments that we're proud of

We're proud of our descriptive graphs and overall conclusions. Despite our limited time on the project, we believe we can make a big impact with the things we learned.

What we learned

We learned about the investments of each state, and how they translate to their overall energy cleanliness. We found that some obvious answers, such as Texas, may have a lot of investments in renewable energy, but also, they proportionally are less "green" than other states.

What's next for Chevron 2023 Datathon

Improving the model and uploading more data is the immediate next step to improve the project. Furthermore, we have the ability to analyze the eco-friendliness of states in more ways and think more critically about their impacts.

Built With

• matplotlib
• pandas
• plotly
• python
• scikit-learn