Classifier combination is a technique that often provides appreciable accuracy gains. In this paper, we argue that the underlying statistical model of classifier combination should be made explicit. Using directed graphical models (DGMs), we provide representations of two common combination schemes, the mean and product rules. We also introduce new DGMs that yield novel combination rules. We find that these new DGM-inspired rules can achieve significant accuracy gains on the TIMIT phone-classification task relative to existing combination schemes.
Cite as: Bilmes, J.A., Kirchhoff, K. (2000) Directed graphical models of classifier combination: application to phone recognition. Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000), vol. 3, 921, doi: 10.21437/ICSLP.2000-684
@inproceedings{bilmes00_icslp,
author={Jeff A. Bilmes and Katrin Kirchhoff},
title={{Directed graphical models of classifier combination: application to phone recognition}},
year=2000,
booktitle={Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000)},
pages={vol. 3, 921},
doi={10.21437/ICSLP.2000-684}
}