Improving classification when a class hierarchy is available using a hierarchy-based prior

DOI10.1214/07-BA209MaRDI QIDQ5965216zbMATH OpenWikidataFDO

Authors Babak Shahbaba, Radford Neal

Publication date 2 March 2016

Published in Bayesian Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/math/0510449, https://projecteuclid.org/euclid.ba/1340390069

zbMATH Keywords

Bayesian modelsdocument classificationhierarchical classificationmultinomial logistic regressionpage layout analysis

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Linear regression; mixed models (62J05)

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 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.

Recommendations

• Classification using hierarchical naïve Bayes models
• Classification using hierarchical Naïve Bayes models
• A survey of hierarchical classification across different application domains
• Large margin hierarchical classification with mutually exclusive class membership
• A top-down hierarchical classification method

Cited in

(4)

• Using natural class hierarchies in multi-class visual classification
• Improving Hierarchical Classification with Partial Labels
• Class-Based Probability Estimation Using a Semantic Hierarchy
• A convergence diagnostic for Bayesian clustering

This page was built for publication: Improving classification when a class hierarchy is available using a hierarchy-based prior

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5965216)