We formulate an axiomatic model to analyze the problem of probability smoothing in Naïve Bayes. We define several desirable properties of smoothing methods. Our main result shows that the Laplace smoothing method is the only one that satisfies ratio preservation , order preservation , and positivity . An alternative characterization based on reallocation-proofness is also obtained.
No takes yet. Share an insight, caveat, or question.
Toyotaka Sakai (2025) studied this question.
Synapse has enriched 2 closely related papers on similar clinical questions. Consider them for comparative context: