0 alternative definitions
A model that defines the probability of a word's presence in a text based on the text that comes before and after it.