Selection of generative models in Classification
Guillaume Bouchard and Gilles Celeux
IEEE Transactions on Pattern Analysis and Macine Intelligence
This article is concerned with the selection of a generative model for supervise
Classical model selection criteria are assessing the fit of a model rather than
its ability to produce
a low classification error rate. A new criterion, the so called Bayesian Entropy
Criterion (BEC) is proposed.
This criterion is taking into account the decisional purpose of a model by minim
izing the integrated classification entropy.
It provides an interesting alternative to
the cross validated error rate which is highly time consuming. The asymptotic be
havior of BEC criterion is presented.
Numerical experiments on both simulated and real data sets show that BEC is perf
orming better than BIC criterion
select a model minimizing the classification error rate and is providing analogo
us performances than the cross validated error rate.
|Additional Information:||Model Selection|
|Project Keyword:||Project Keyword UNSPECIFIED|
|Subjects:||Theory & Algorithms|
|Deposited By:||Gilles Celeux|
|Deposited On:||29 November 2005|