On the optimality of the simple Bayesian classi er under zero-one loss (1997)

by Pedro Domingos, Michael Pazzani
Venue:Machine Learning