Open Access
Translator Disclaimer
March 2007 Improving classification when a class hierarchy is available using a hierarchy-based prior
Radford M. Neal, Babak Shahbaba
Bayesian Anal. 2(1): 221-237 (March 2007). DOI: 10.1214/07-BA209

Abstract

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 a different way. We also test the new method on page layout analysis and document classification problems, and find that it performs better than the other methods.

Citation

Download Citation

Radford M. Neal. Babak Shahbaba. "Improving classification when a class hierarchy is available using a hierarchy-based prior." Bayesian Anal. 2 (1) 221 - 237, March 2007. https://doi.org/10.1214/07-BA209

Information

Published: March 2007
First available in Project Euclid: 22 June 2012

zbMATH: 1331.62316
MathSciNet: MR2289929
Digital Object Identifier: 10.1214/07-BA209

Subjects:
Primary: Database Expansion Item

Keywords: Bayesian models , Document Classification , Hierarchical Classification , Multinomial Logistic Regression , Page Layout Analysis

Rights: Copyright © 2007 International Society for Bayesian Analysis

JOURNAL ARTICLE
17 PAGES


SHARE
Vol.2 • No. 1 • March 2007
Back to Top