We propose a multistage recognition method built as a cascade of a linear parametric model and a $k$-nearest neighbor ($k$-NN) nonparametric classifier. The linear model learns a “rule” and the $k$-NN learns the “exceptions” rejected by the “rule.” Because the rule-learner handles a large percentage of the examples using a simple and general rule, only a small subset of the training set is stored as exceptions during training. Similarly during testing, most patterns are handled by the rule -learner and few are handled by the exception-learner thus causing only a small increase in memory and computation. A multistage method like cascading is a better approach than a multiexpert method like voting where all learners are used for all cases; the extra computation and memory for the second learner is unnecessary if we are sufficiently certain that the first one’s response is correct. We discuss how such a system can be trained using cross validation. This method is tested on the real-world application of handwritten digit recognition.
We propose a multistage recognition method built as a cascade of a linear parametric model and a $k$-nearest neighbor ($k$-NN) nonparametric classifier. The linear model learns a “rule” and the $k$-NN learns the “exceptions” rejected by the “rule.” Because the rule-learner handles a large percentage of the examples using a simple and general rule, only a small subset of the training set is stored as exceptions during training. Similarly during testing, most patterns are handled by the rule -learner and few are handled by the exception-learner thus causing only a small increase in memory and computation. A multistage method like cascading is a better approach than a multiexpert method like voting where all learners are used for all cases; the extra computation and memory for the second learner is unnecessary if we are sufficiently certain that the first one’s response is correct. We discuss how such a system can be trained using cross validation. This method is tested on the real-world application of handwritten digit recognition.
@article{KYB_1998_34_4_a2,
author = {Alpaydin, Ethem and Kaynak, Cenk},
title = {Cascading classifiers},
journal = {Kybernetika},
pages = {369--374},
year = {1998},
volume = {34},
number = {4},
zbl = {1274.68284},
language = {en},
url = {http://geodesic.mathdoc.fr/item/KYB_1998_34_4_a2/}
}
TY - JOUR
AU - Alpaydin, Ethem
AU - Kaynak, Cenk
TI - Cascading classifiers
JO - Kybernetika
PY - 1998
SP - 369
EP - 374
VL - 34
IS - 4
UR - http://geodesic.mathdoc.fr/item/KYB_1998_34_4_a2/
LA - en
ID - KYB_1998_34_4_a2
ER -
[1] Alpaydın E.: 1997. REx: Learning A Rule and Exceptions. International Computer Science Institute TR-97-040 Berkeley
[2] Alpaydın E., Gürgen F.: Comparison of kernel estimators, perceptrons and radial–basis functions for OCR and speech classification. Neural Computing Appl. 3 (1995), 38–49 | DOI
[3] Bishop C. M.: Neural Networks for Pattern Recognition. Oxford University Press, Oxford 1995 | MR | Zbl
[4] Garris M. D., Blue J. L., Candela G. T., Dimmick D. L., Geist J., Grother P. J., Janet S. A., Wilson C. L.: NIST Form–Based Handprint Recognition System, NISTIR 5469, 199.
[5] Pudil P., Novovičová J., Bláha S., Kittler J.: Multistage pattern recognition with reject option. In: 11th IAPR International Conference on Pattern Recognition B, 1992, vol. II, pp. 92–95
[6] Xu L., Krzyżak, A., Suen C. Y.: Methods of combining multiple classifiers and their applications to handwriting recognition. IEEE Trans. Systems Man Cybernet. 22 (1992), 418–435 | DOI
