Piecewise linear classifiers preserving high local recognition rates

Hiroshi Tenmoto; Mineichi Kudo; Masaru Shimbo

Kybernetika (1998)

  • Volume: 34, Issue: 4, page [479]-484
  • ISSN: 0023-5954

Abstract

top
We propose a new method to construct piecewise linear classifiers. This method constructs hyperplanes of a piecewise linear classifier so as to keep the correct recognition rate over a threshold for a training set. The threshold is determined automatically by the MDL (Minimum Description Length) criterion so as to avoid overfitting of the classifier to the training set. The proposed method showed better results in some experiments than a previous method.

How to cite

top

Tenmoto, Hiroshi, Kudo, Mineichi, and Shimbo, Masaru. "Piecewise linear classifiers preserving high local recognition rates." Kybernetika 34.4 (1998): [479]-484. <http://eudml.org/doc/33381>.

@article{Tenmoto1998,
abstract = {We propose a new method to construct piecewise linear classifiers. This method constructs hyperplanes of a piecewise linear classifier so as to keep the correct recognition rate over a threshold for a training set. The threshold is determined automatically by the MDL (Minimum Description Length) criterion so as to avoid overfitting of the classifier to the training set. The proposed method showed better results in some experiments than a previous method.},
author = {Tenmoto, Hiroshi, Kudo, Mineichi, Shimbo, Masaru},
journal = {Kybernetika},
keywords = {piecewise linear classifiers; clustering; recognition rates; piecewise linear classifiers; clustering; recognition rates},
language = {eng},
number = {4},
pages = {[479]-484},
publisher = {Institute of Information Theory and Automation AS CR},
title = {Piecewise linear classifiers preserving high local recognition rates},
url = {http://eudml.org/doc/33381},
volume = {34},
year = {1998},
}

TY - JOUR
AU - Tenmoto, Hiroshi
AU - Kudo, Mineichi
AU - Shimbo, Masaru
TI - Piecewise linear classifiers preserving high local recognition rates
JO - Kybernetika
PY - 1998
PB - Institute of Information Theory and Automation AS CR
VL - 34
IS - 4
SP - [479]
EP - 484
AB - We propose a new method to construct piecewise linear classifiers. This method constructs hyperplanes of a piecewise linear classifier so as to keep the correct recognition rate over a threshold for a training set. The threshold is determined automatically by the MDL (Minimum Description Length) criterion so as to avoid overfitting of the classifier to the training set. The proposed method showed better results in some experiments than a previous method.
LA - eng
KW - piecewise linear classifiers; clustering; recognition rates; piecewise linear classifiers; clustering; recognition rates
UR - http://eudml.org/doc/33381
ER -

References

top
  1. Forgy E. W., Cluster analysis of multivariate data: Efficiency versus interpretability of classifications, Abstracts Biometrics 21 (1965), 3, 768 (1965) 
  2. al S. Hayamizu et, Generation of VCV/CVC balanced word sets for speech database, Bull. Electrotechnical Laboratory 49 (1985), 10, 803–834 (1985) 
  3. Mangasarian O. L., 10.1109/TIT.1968.1054229, IEEE Trans. Inform. Theory 14 (1968), 6, 801–807 (1968) Zbl0169.51108DOI10.1109/TIT.1968.1054229
  4. Meisel W. S., Computer–Oriented Approaches to Pattern Recognition, Academic Press, New York 1972 Zbl0252.68063
  5. Nilsson N. J., Learning Machines, McGraw–Hill, New York 1965 Zbl0760.68060
  6. Park Y., Sklansky J., 10.1007/BF01908599, J. Classification 6 (1989), 195–222 (1989) Zbl0698.62061DOI10.1007/BF01908599
  7. Rissanen J., 10.1214/aos/1176346150, Ann. Statist. 11 (1983), 416–431 (1983) Zbl0513.62005DOI10.1214/aos/1176346150
  8. Sklansky J., Michelotti L., 10.1109/TPAMI.1980.4766988, IEEE Trans. Pattern Analysis and Machine Intelligence 2 (1980), 101–111 (1980) Zbl0443.62046DOI10.1109/TPAMI.1980.4766988
  9. Sklansky J., Wassel G. N., Pattern Classification and Trainable Machines, Springer–Verlag, New York 1981 
  10. Takiyama R., 10.1016/0031-3203(80)90005-9, Pattern Recognition 12 (1980), 75–82 (1980) Zbl0428.68092MR0567072DOI10.1016/0031-3203(80)90005-9
  11. Tomek I., 10.1109/TSMC.1976.4309452, IEEE Trans. Systems Man Cybernet. 6 (1976), 11, 769–772 (1976) Zbl0341.68066MR0449068DOI10.1109/TSMC.1976.4309452
  12. Yamanishi K., A learning criterion for stochastic rules, In: Machine Learning, to appear. An extended abstract in Proceedings of the Third Annual Workshop on Computational Learning Theory, 1990, pp. 67–81 (1990) 

NotesEmbed ?

top

You must be logged in to post comments.

To embed these notes on your page include the following JavaScript code on your page where you want the notes to appear.

Only the controls for the widget will be shown in your chosen language. Notes will be shown in their authored language.

Tells the widget how many notes to show per page. You can cycle through additional notes using the next and previous controls.

    
                

                
Note: Best practice suggests putting the JavaScript code just before the closing </body> tag.