Empirical Machine Learning Comparison
This paper was done as a final project for my COGS 118A: Supervised Machine Learning Algorithms class. I use a machine learning pipeline on four datasets taken from the UCI machine learning data repository and create an empirical comparison between various ML algorithms using the Caruana paper from 2006 as a model for how the experiments were to be conducted. More details are available in the paper as to how I conducted my experiments. I used a pipeline for each different dataset and ran each dataset separately because it was easier to bug fix that way in a timely manner (specifically before the deadline of the project). The miscellaneous calculations notebook was used to compile averages and metrics across datasets, which could not be done in the other notebooks because of the way I formatted it.