Linguistic Modelling Using a semi-naive Bayes Framework
NJ Randon, Jonathan Lawry · Bristol Research (University of Bristol) · 2002
A random set semantics is presented as a knowledge representation framework for learning linguistic prototypes. Within this framework a number of algorithms for learning prototypes are presented, based on grouping certain sets of attributes and evaluating joint mass assignments on labels. Such prototypes are then combined with a semi-Naive Bayes classifier in order to determine classification probabilities. The potential of such linguistic classifiers is then illustrated by their application to a number of toy and benchmark problems.