14th IEEE Symposium on Computer-Based Medical Systems (CMBS'01)
Ensemble Classification of the VF Dataset with Limited Merging
Bethesda, Maryland
March 26-March 27
ISBN: 0-7695-1004-3
Abstract: Ensemble classification (the concurrent development of K independent classifications) is a common practice in Gibbs classification. Here we describe a adaptation of Azencott's theorem on finite-time annealing, coupled with a limited merging protocol, which produces a final low-energy Gibbs classification of the VF dataset.
Index Terms:
Gibbs classification, finite-time annealing, Visible Human Dataset.
Citation:
Zhihong Yang, Ian R. Greenshields, "Ensemble Classification of the VF Dataset with Limited Merging," cbms, pp.0117, 14th IEEE Symposium on Computer-Based Medical Systems (CMBS'01), 2001