Applied Semantic Web Technologies

266  ◾  Dae-Ki Kang

attributes. ese taxonomy learning algorithms recursively group values based on

a suitable measure of divergence between the class distributions associated with

the values to construct taxonomies. ey can generate hierarchical taxonomies of

nominal, ordinal, and continuous valued attributes.

For evaluation of the generated taxonomies, we described taxonomy-aware

machine learning algorithms such as AVT-NBL, a generalization of the naive Bayes

learner for multivariate data using AVTs, WTNBL-MN, a generalization of the

naive Bayes learner for the multinomial event model for learning classiers from

data using word taxonomy, PAT-NBL, a generalization of the naive Bayes learner

to exploit propositionalized attribute ta