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

by Pedro Domingos, Michael Pazzani
Venue:MACHINE LEARNING