Improving Classification When a Class Hierarchy is Available Using a Hierarchy-Based Prior

Babak Shahbaba, Dept. of Public Health Sciences, University of Toronto
Radford M. Neal, Dept. of Statistics and Dept. of Computer Science, University of Toronto

We introduce a new method for building classification models when we have prior knowledge of how the classes can be arranged in a hierarchy, based on how easily they can be distinguished. The new method uses a Bayesian form of the multinomial logit (MNL, a.k.a. ``softmax'') model, with a prior that introduces correlations between the parameters for classes that are nearby in the tree. We compare the performance on simulated data of the new method, the ordinary MNL model, and a model that uses the hierarchy in different way. We also test the new method on a document labelling problem, and find that it performs better than the other methods, particularly when the amount of training data is small.

Technical Report No. 0510, Dept. of Statistics, University of Toronto (October 2005), 11 pages: postscript, pdf.

Also available from arXiv.org.

Associated references: The following technical report applies the models introduced here to gene function classification:
Shahbaba, B. and Neal, R. M. (2006) ``Gene function classification using Bayesian models with hierarchy-based priors'', Technical Report No. 0606, Dept. of Statistics, 14 pages: abstract, postscript, pdf, associated references.