Leave-one-out Evaluation of the Nearest Feature Line and the Rectified Nearest Feature Line Segment Classifiers Using Multi-core Architectures

In this paper we present the parallelization of the leave-one-out test: a reproducible test that is, in general, computationally expensive. Parallelization was implemented on multi-core multi-threaded architectures, using the Flynn Single Instruction Multiple Data taxonomy. This technique was used f...

Full description

Autores:
Uribe-Hurtado, Ana Lorena
Villegas-Jaramillo, Eduardo-Jose
Orozco-Alzate, Mauricio
Tipo de recurso:
Fecha de publicación:
2018
Institución:
Universidad EAFIT
Repositorio:
Repositorio EAFIT
Idioma:
eng
OAI Identifier:
oai:repository.eafit.edu.co:10784/13192
Acceso en línea:
http://publicaciones.eafit.edu.co/index.php/ingciencia/article/view/4867
http://hdl.handle.net/10784/13192
Palabra clave:
Multi-core computing
Classification algorithms
Leave-one-out test
Computación con múltiples núcleos
Algoritmos de clasificación
Prueba leave-one-ou
Rights
License
Attribution 4.0 International (CC BY 4.0)
id REPOEAFIT2_3c0a2534080279c51265380c101c66a3
oai_identifier_str oai:repository.eafit.edu.co:10784/13192
network_acronym_str REPOEAFIT2
network_name_str Repositorio EAFIT
repository_id_str
dc.title.eng.fl_str_mv Leave-one-out Evaluation of the Nearest Feature Line and the Rectified Nearest Feature Line Segment Classifiers Using Multi-core Architectures
dc.title.spa.fl_str_mv Evaluación leave-one-out de los clasificadores de la línea de características más cercana y del segmento de línea rectificado más cercano usando arquitecturas multi-núcleo
title Leave-one-out Evaluation of the Nearest Feature Line and the Rectified Nearest Feature Line Segment Classifiers Using Multi-core Architectures
spellingShingle Leave-one-out Evaluation of the Nearest Feature Line and the Rectified Nearest Feature Line Segment Classifiers Using Multi-core Architectures
Multi-core computing
Classification algorithms
Leave-one-out test
Computación con múltiples núcleos
Algoritmos de clasificación
Prueba leave-one-ou
title_short Leave-one-out Evaluation of the Nearest Feature Line and the Rectified Nearest Feature Line Segment Classifiers Using Multi-core Architectures
title_full Leave-one-out Evaluation of the Nearest Feature Line and the Rectified Nearest Feature Line Segment Classifiers Using Multi-core Architectures
title_fullStr Leave-one-out Evaluation of the Nearest Feature Line and the Rectified Nearest Feature Line Segment Classifiers Using Multi-core Architectures
title_full_unstemmed Leave-one-out Evaluation of the Nearest Feature Line and the Rectified Nearest Feature Line Segment Classifiers Using Multi-core Architectures
title_sort Leave-one-out Evaluation of the Nearest Feature Line and the Rectified Nearest Feature Line Segment Classifiers Using Multi-core Architectures
dc.creator.fl_str_mv Uribe-Hurtado, Ana Lorena
Villegas-Jaramillo, Eduardo-Jose
Orozco-Alzate, Mauricio
dc.contributor.author.none.fl_str_mv Uribe-Hurtado, Ana Lorena
Villegas-Jaramillo, Eduardo-Jose
Orozco-Alzate, Mauricio
dc.contributor.affiliation.spa.fl_str_mv Universidad Nacional de Colombia sede Manizales
Universidad Nacional de Colombia
dc.subject.keyword.eng.fl_str_mv Multi-core computing
Classification algorithms
Leave-one-out test
topic Multi-core computing
Classification algorithms
Leave-one-out test
Computación con múltiples núcleos
Algoritmos de clasificación
Prueba leave-one-ou
dc.subject.keyword.spa.fl_str_mv Computación con múltiples núcleos
Algoritmos de clasificación
Prueba leave-one-ou
description In this paper we present the parallelization of the leave-one-out test: a reproducible test that is, in general, computationally expensive. Parallelization was implemented on multi-core multi-threaded architectures, using the Flynn Single Instruction Multiple Data taxonomy. This technique was used for the preprocessing and processing stages of two classification algorithms that are oriented to enrich the representation in small sample cases: the nearest feature line (NFL) algorithm and the rectified nearest feature line segment (RNFLS) algorithm. Results show an acceleration of up to 18.17 times with the smallest dataset and 29.91 times with the largest one, using the most costly algorithm (RNFLS) whose complexity is O(n4). The paper also shows the pseudo-codes of the serial and parallel algorithms using, in the latter case, a notation that describes the way the parallelization was carried out as a function of the threads.
publishDate 2018
dc.date.available.none.fl_str_mv 2018-11-16T16:28:59Z
dc.date.issued.none.fl_str_mv 2018-06-14
dc.date.accessioned.none.fl_str_mv 2018-11-16T16:28:59Z
dc.date.none.fl_str_mv 2018-06-14
dc.type.eng.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
article
publishedVersion
dc.type.coarversion.fl_str_mv http://purl.org/coar/version/c_970fb48d4fbd8a85
dc.type.coar.fl_str_mv http://purl.org/coar/resource_type/c_6501
http://purl.org/coar/resource_type/c_2df8fbb1
dc.type.local.spa.fl_str_mv Artículo
status_str publishedVersion
dc.identifier.issn.none.fl_str_mv 2256-4314
1794-9165
dc.identifier.uri.none.fl_str_mv http://publicaciones.eafit.edu.co/index.php/ingciencia/article/view/4867
http://hdl.handle.net/10784/13192
dc.identifier.doi.none.fl_str_mv 10.17230/ingciencia.13.27.4
identifier_str_mv 2256-4314
1794-9165
10.17230/ingciencia.13.27.4
url http://publicaciones.eafit.edu.co/index.php/ingciencia/article/view/4867
http://hdl.handle.net/10784/13192
dc.language.iso.none.fl_str_mv eng
language eng
dc.relation.isversionof.none.fl_str_mv http://publicaciones.eafit.edu.co/index.php/ingciencia/article/view/4867
dc.rights.eng.fl_str_mv Attribution 4.0 International (CC BY 4.0)
dc.rights.coar.fl_str_mv http://purl.org/coar/access_right/c_abf2
dc.rights.uri.none.fl_str_mv http://creativecommons.org/licenses/by/4.0
dc.rights.local.spa.fl_str_mv Acceso abierto
rights_invalid_str_mv Attribution 4.0 International (CC BY 4.0)
http://creativecommons.org/licenses/by/4.0
Acceso abierto
http://purl.org/coar/access_right/c_abf2
dc.format.none.fl_str_mv application/pdf
dc.publisher.spa.fl_str_mv Universidad EAFIT
dc.source.none.fl_str_mv instname:Universidad EAFIT
reponame:Repositorio Institucional Universidad EAFIT
dc.source.eng.fl_str_mv Ingeniería y Ciencia; Vol 14 No 27 (2018); 75-99
dc.source.spa.fl_str_mv Ingeniería y Ciencia; Vol 14 No 27 (2018); 75-99
instname_str Universidad EAFIT
institution Universidad EAFIT
reponame_str Repositorio Institucional Universidad EAFIT
collection Repositorio Institucional Universidad EAFIT
bitstream.url.fl_str_mv https://repository.eafit.edu.co/bitstreams/d99a7839-a7b9-4bd9-b48c-5b21ae224f66/download
https://repository.eafit.edu.co/bitstreams/e981cd01-488c-464e-b85b-4aa5679caaf9/download
https://repository.eafit.edu.co/bitstreams/e4dbcc43-0eef-430b-840e-772737ecdea0/download
bitstream.checksum.fl_str_mv da9b21a5c7e00c7f1127cef8e97035e0
2166f85e510370a730ca215aba4b28c0
e05d441677441e6293db0f1a57caf5e4
bitstream.checksumAlgorithm.fl_str_mv MD5
MD5
MD5
repository.name.fl_str_mv Repositorio Institucional Universidad EAFIT
repository.mail.fl_str_mv repositorio@eafit.edu.co
_version_ 1814110324161249280
spelling 2018-06-142018-11-16T16:28:59Z2018-06-142018-11-16T16:28:59Z2256-43141794-9165http://publicaciones.eafit.edu.co/index.php/ingciencia/article/view/4867http://hdl.handle.net/10784/1319210.17230/ingciencia.13.27.4In this paper we present the parallelization of the leave-one-out test: a reproducible test that is, in general, computationally expensive. Parallelization was implemented on multi-core multi-threaded architectures, using the Flynn Single Instruction Multiple Data taxonomy. This technique was used for the preprocessing and processing stages of two classification algorithms that are oriented to enrich the representation in small sample cases: the nearest feature line (NFL) algorithm and the rectified nearest feature line segment (RNFLS) algorithm. Results show an acceleration of up to 18.17 times with the smallest dataset and 29.91 times with the largest one, using the most costly algorithm (RNFLS) whose complexity is O(n4). The paper also shows the pseudo-codes of the serial and parallel algorithms using, in the latter case, a notation that describes the way the parallelization was carried out as a function of the threads.Presentamos en este artículo la paralelización de la prueba leave-one-out, la cual es una prueba repetible pero que, en general, resulta costosa computacionalmente. La paralelización se implementó sobre arquitecturas multinúcleo con múltiples hilos, usando la taxonomía Flynn Single Instruction Multiple Data. Esta técnica se empleó para las etapas de preproceso y proceso de dos algoritmos de clasificación que están orientados a enriquecer la representación en casos de muestra pequeña: el algoritmo de la línea de características más cercana (NFL) y el algoritmo del segmento de línea rectificado más cercano (RNFLS). Los resultados obtenidos muestran una aceleración de hasta 18.17 veces con el conjunto de datos mas pequeño y de 29.91 veces con el conjunto de datos más grande, empleando el algoritmo más costoso —RNFLS— cuya complejidad es O(n4). El artículo muestra también los pseudocódigos de los algoritmos seriales y paralelos empleando, en este último caso, una notación que describe la manera como se realizó la paralelización en función de los hilos.application/pdfengUniversidad EAFIThttp://publicaciones.eafit.edu.co/index.php/ingciencia/article/view/4867Copyright (c) 2018 Ana Lorena Uribe-Hurtado, Eduardo-Jose Villegas-Jaramillo, Mauricio Orozco-AlzateAttribution 4.0 International (CC BY 4.0)http://creativecommons.org/licenses/by/4.0Acceso abiertohttp://purl.org/coar/access_right/c_abf2instname:Universidad EAFITreponame:Repositorio Institucional Universidad EAFITIngeniería y Ciencia; Vol 14 No 27 (2018); 75-99Ingeniería y Ciencia; Vol 14 No 27 (2018); 75-99Leave-one-out Evaluation of the Nearest Feature Line and the Rectified Nearest Feature Line Segment Classifiers Using Multi-core ArchitecturesEvaluación leave-one-out de los clasificadores de la línea de características más cercana y del segmento de línea rectificado más cercano usando arquitecturas multi-núcleoinfo:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionarticlepublishedVersionArtículohttp://purl.org/coar/version/c_970fb48d4fbd8a85http://purl.org/coar/resource_type/c_6501http://purl.org/coar/resource_type/c_2df8fbb1Multi-core computingClassification algorithmsLeave-one-out testComputación con múltiples núcleosAlgoritmos de clasificaciónPrueba leave-one-ouUribe-Hurtado, Ana LorenaVillegas-Jaramillo, Eduardo-JoseOrozco-Alzate, MauricioUniversidad Nacional de Colombia sede ManizalesUniversidad Nacional de ColombiaIngeniería y Ciencia14277599ing.ciencTHUMBNAILminaitura-ig_Mesa de trabajo 1.jpgminaitura-ig_Mesa de trabajo 1.jpgimage/jpeg265796https://repository.eafit.edu.co/bitstreams/d99a7839-a7b9-4bd9-b48c-5b21ae224f66/downloadda9b21a5c7e00c7f1127cef8e97035e0MD51ORIGINALdocument (4).pdfdocument (4).pdfTexto completo PDFapplication/pdf826697https://repository.eafit.edu.co/bitstreams/e981cd01-488c-464e-b85b-4aa5679caaf9/download2166f85e510370a730ca215aba4b28c0MD52articulo.htmlarticulo.htmlTexto completo HTMLtext/html374https://repository.eafit.edu.co/bitstreams/e4dbcc43-0eef-430b-840e-772737ecdea0/downloade05d441677441e6293db0f1a57caf5e4MD5310784/13192oai:repository.eafit.edu.co:10784/131922020-03-01 12:48:13.032http://creativecommons.org/licenses/by/4.0Copyright (c) 2018 Ana Lorena Uribe-Hurtado, Eduardo-Jose Villegas-Jaramillo, Mauricio Orozco-Alzateopen.accesshttps://repository.eafit.edu.coRepositorio Institucional Universidad EAFITrepositorio@eafit.edu.co