Comparison of two ML models for Classification using U.C.I. Car Evaluation Data set
Machine Learning models applied to the car evaluation data can provide clear commercial direction and competition-benefit to manufacturer’s since desirable features can be embedded in the future design process. Herein we compare a Naive Bayes approach with a bagged decision tree model which extends to a random forest model upon hyper-parameter tuning. We critically assess our results with those of W. Piraya and Behzad, whom also built classification models for this same car evaluation data-set.
Author: Harry Li, Paul O’Donovan