Naive Bayes Classifiers and One-hot Encoding of Categorical Variables

arXiv:2404.18190v1 Announce Type: new
Abstract: This paper investigates the consequences of encoding a $K$-valued categorical variable incorrectly as $K$ bits via one-hot encoding, when using a Na\"{\i}ve Bayes classifier. This gives rise to a product-of-Bernoullis (PoB) assumption, rather than the correct categorical Na\"{\i}ve Bayes classifier. The differences between the two classifiers are analysed mathematically and experimentally. In our experiments using probability vectors drawn from a Dirichlet distribution, the two classifiers are found to agree on the maximum a posteriori class …

abstract arxiv bayes categorical classifier classifiers consequences cs.lg differences encoding hot paper product stat.ml type variables via