This is a poster I made as part of the final project for CS229: Machine Learning. For this project, we decided to apply machine learning to sports betting by designing models to predict events that will happen in an NBA game. We decided to focus on the Over/Under, the total number of points scored by both teams combined.

I think this poster reflects well what I learned from my poster in CS221. We decided to stick with the red colors to build ethos as Stanford researchers, and improved upon the poster structure that I had in CS221. I think we were able to tell a clearer story of our project, and explain the motivation and impacts of our work. The fact that our poster organization closely follows that of a computer science research paper helped us present our work more clearly to our expert audience: the teaching staff and our classmates. This organization choice also helped affirm our credibility.

While we didn’t use many visuals and images on this poster, I think it still effectively communicates all the necessary information. One section that could maybe use another visual is the data section: it might have been helpful to include a visualization of our dataset or at least a snapshot of a training example. We could have also been more creative in the way we presented our results, as we could have included more plots of the many metrics we used to evaluate our models. In the end, we moved away from these two options because we felt like we needed more space to discuss the impact of our results, especially since they relate to a new industry: sports betting. We also had to accommodate the poster requirements for the class.

If I were to redesign this poser today, I would think more about how to remove some of the text that we had to replace it with either more whitespace or more visuals, as this poster is very crowded.

This project was done in collaboration with Vishnu Sarukkai.