Using maximum similarity graphs to edit nearest neighbor classifiers

Milton García-Borroto; Yenny Villuendas-Rey; Jesús Ariel Carrasco-Ochoa; José Fco Martínez-Trinidad

doi:10.1007/978-3-642-10268-4_57

Using maximum similarity graphs to edit nearest neighbor classifiers

Milton García-Borroto, Yenny Villuendas-Rey, Jesús Ariel Carrasco-Ochoa, José Fco Martínez-Trinidad

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

10 Scopus citations

Abstract

The Nearest Neighbor classifier is a simple but powerful nonparametric technique for supervised classification. However, it is very sensitive to noise and outliers, which could decrease the classifier accuracy. To overcome this problem, we propose two new editing methods based on maximum similarity graphs. Numerical experiments in several databases show the high quality performance of our methods according to classifier accuracy.

Original language	English
Title of host publication	Progress in Pattern Recognition, Image Analysis, Computer Vision and Applications - 14th Iberoamerican Conference on Pattern Recognition, CIARP 2009, Proceedings
Pages	489-496
Number of pages	8
DOIs	https://doi.org/10.1007/978-3-642-10268-4_57
State	Published - 2009
Externally published	Yes
Event	14th Iberoamerican Conference on Pattern Recognition, CIARP 2009 - Guadalajara, Jalisco, Mexico Duration: 15 Nov 2009 → 18 Nov 2009

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	5856 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	14th Iberoamerican Conference on Pattern Recognition, CIARP 2009
Country/Territory	Mexico
City	Guadalajara, Jalisco
Period	15/11/09 → 18/11/09

Keywords

Error-based editing
Nearest neighbor
Prototype selection

Access to Document

10.1007/978-3-642-10268-4_57

Cite this

García-Borroto, M., Villuendas-Rey, Y., Carrasco-Ochoa, J. A., & Martínez-Trinidad, J. F. (2009). Using maximum similarity graphs to edit nearest neighbor classifiers. In Progress in Pattern Recognition, Image Analysis, Computer Vision and Applications - 14th Iberoamerican Conference on Pattern Recognition, CIARP 2009, Proceedings (pp. 489-496). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5856 LNCS). https://doi.org/10.1007/978-3-642-10268-4_57

García-Borroto, Milton ; Villuendas-Rey, Yenny ; Carrasco-Ochoa, Jesús Ariel et al. / Using maximum similarity graphs to edit nearest neighbor classifiers. Progress in Pattern Recognition, Image Analysis, Computer Vision and Applications - 14th Iberoamerican Conference on Pattern Recognition, CIARP 2009, Proceedings. 2009. pp. 489-496 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{841d72c887084b5b9dbfde5e3d1356a2,

title = "Using maximum similarity graphs to edit nearest neighbor classifiers",

abstract = "The Nearest Neighbor classifier is a simple but powerful nonparametric technique for supervised classification. However, it is very sensitive to noise and outliers, which could decrease the classifier accuracy. To overcome this problem, we propose two new editing methods based on maximum similarity graphs. Numerical experiments in several databases show the high quality performance of our methods according to classifier accuracy.",

keywords = "Error-based editing, Nearest neighbor, Prototype selection",

author = "Milton Garc{\'i}a-Borroto and Yenny Villuendas-Rey and Carrasco-Ochoa, {Jes{\'u}s Ariel} and Mart{\'i}nez-Trinidad, {Jos{\'e} Fco}",

year = "2009",

doi = "10.1007/978-3-642-10268-4_57",

language = "Ingl{\'e}s",

isbn = "3642102670",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "489--496",

booktitle = "Progress in Pattern Recognition, Image Analysis, Computer Vision and Applications - 14th Iberoamerican Conference on Pattern Recognition, CIARP 2009, Proceedings",

note = "14th Iberoamerican Conference on Pattern Recognition, CIARP 2009 ; Conference date: 15-11-2009 Through 18-11-2009",

}

García-Borroto, M, Villuendas-Rey, Y, Carrasco-Ochoa, JA & Martínez-Trinidad, JF 2009, Using maximum similarity graphs to edit nearest neighbor classifiers. in Progress in Pattern Recognition, Image Analysis, Computer Vision and Applications - 14th Iberoamerican Conference on Pattern Recognition, CIARP 2009, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5856 LNCS, pp. 489-496, 14th Iberoamerican Conference on Pattern Recognition, CIARP 2009, Guadalajara, Jalisco, Mexico, 15/11/09. https://doi.org/10.1007/978-3-642-10268-4_57

Using maximum similarity graphs to edit nearest neighbor classifiers. / García-Borroto, Milton; Villuendas-Rey, Yenny; Carrasco-Ochoa, Jesús Ariel et al.
Progress in Pattern Recognition, Image Analysis, Computer Vision and Applications - 14th Iberoamerican Conference on Pattern Recognition, CIARP 2009, Proceedings. 2009. p. 489-496 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5856 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Using maximum similarity graphs to edit nearest neighbor classifiers

AU - García-Borroto, Milton

AU - Villuendas-Rey, Yenny

AU - Carrasco-Ochoa, Jesús Ariel

AU - Martínez-Trinidad, José Fco

PY - 2009

Y1 - 2009

N2 - The Nearest Neighbor classifier is a simple but powerful nonparametric technique for supervised classification. However, it is very sensitive to noise and outliers, which could decrease the classifier accuracy. To overcome this problem, we propose two new editing methods based on maximum similarity graphs. Numerical experiments in several databases show the high quality performance of our methods according to classifier accuracy.

AB - The Nearest Neighbor classifier is a simple but powerful nonparametric technique for supervised classification. However, it is very sensitive to noise and outliers, which could decrease the classifier accuracy. To overcome this problem, we propose two new editing methods based on maximum similarity graphs. Numerical experiments in several databases show the high quality performance of our methods according to classifier accuracy.

KW - Error-based editing

KW - Nearest neighbor

KW - Prototype selection

UR - http://www.scopus.com/inward/record.url?scp=78651261630&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-10268-4_57

DO - 10.1007/978-3-642-10268-4_57

M3 - Contribución a la conferencia

SN - 3642102670

SN - 9783642102677

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 489

EP - 496

BT - Progress in Pattern Recognition, Image Analysis, Computer Vision and Applications - 14th Iberoamerican Conference on Pattern Recognition, CIARP 2009, Proceedings

T2 - 14th Iberoamerican Conference on Pattern Recognition, CIARP 2009

Y2 - 15 November 2009 through 18 November 2009

ER -

García-Borroto M, Villuendas-Rey Y, Carrasco-Ochoa JA, Martínez-Trinidad JF. Using maximum similarity graphs to edit nearest neighbor classifiers. In Progress in Pattern Recognition, Image Analysis, Computer Vision and Applications - 14th Iberoamerican Conference on Pattern Recognition, CIARP 2009, Proceedings. 2009. p. 489-496. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-642-10268-4_57

Using maximum similarity graphs to edit nearest neighbor classifiers

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this