I completed this project for a Data Science class at UT Austin. This project uses various environmental and human factors to predict pollution levels in regions across the US. The results are then compared to the actual measured pollution levels. The main idea of this project is that if we are able to reliably estimate pollution levels, the need for expensive monitoring equipment could be mitigated. Areas with a lower budget that are disproportionately affected by pollution that don’t monitor as well could also have a better/cheaper way to measure their pollution. The result of testing many different models showed that the “RF” model preformed the best with an RMSE of ~1.66 which is very good.