Previous |  Up |  Next

Article

Title: Decomposition of high dimensional pattern spaces for hierarchical classification (English)
Author: Kumar, Rajeev
Author: Rockett, Peter
Language: English
Journal: Kybernetika
ISSN: 0023-5954
Volume: 34
Issue: 4
Year: 1998
Pages: [435]-442
Summary lang: English
.
Category: math
.
Summary: In this paper we present a novel approach to decomposing high dimensional spaces using a multiobjective genetic algorithm for identifying (near-)optimal subspaces for hierarchical classification. This strategy of pre-processing the data and explicitly optimising the partitions for subsequent mapping onto a hierarchical classifier is found to both reduce the learning complexity and the classification time with no degradation in overall classification error rate. Results of partitioning pattern spaces are presented and compared with various algorithms. (English)
Keyword: pre-processing
Keyword: decomposition
Keyword: pattern classifiers
MSC: 68T05
MSC: 68T10
idZBL: Zbl 1274.68329
.
Date available: 2009-09-24T19:18:39Z
Last updated: 2015-03-28
Stable URL: http://hdl.handle.net/10338.dmlcz/135228
.
Reference: [1] Friedman J. H.: A recursive partitioning decision role for nonparametric classification.IEEE Trans. Comput. 26 (1997), 4, 404–408
Reference: [2] Henrichon E. G., Fu K. S.: A nonparametric partitioning procedure for pattern classification.IEEE Trans. Comput. 18 (1969), 7, 614–624 10.1109/T-C.1969.222728
Reference: [3] Kanal L. N.: Problem–solving models and search strategies for pattern recognition, IEEE Trans.Pattern Analysis Machine Intelligence 1 (1979), 2, 193–201 10.1109/TPAMI.1979.4766905
Reference: [4] Kumar R.: Feature Selection, Representation and Classification in Vision.Ph.D. Thesis, Dept. Electronic and Electrical Engineering, University of Sheffield, 1997
Reference: [5] al C. C. Taylor et: Dataset descriptions and results.In: Machine Learning, Neural and Statistical Classification (D. Michie, D. J. Spiegelhalter and C. C. Taylor, eds.), Ellis Horwood, London 1994, pp. 131–174
.

Files

Files Size Format View
Kybernetika_34-1998-4_14.pdf 1.125Mb application/pdf View/Open
Back to standard record
Partner of
EuDML logo