Show simple item record

dc.contributor.authorGonzález Acuña, Hernánspa
dc.contributor.authorDutra, Max Suellspa
dc.contributor.authorLengerke, Omarspa
dc.contributor.authorMorales, Magda
dc.description.abstractEn este artículo se presenta una demostración de los mapas auto organizados de Kohonen's también llamados SOM. Así mismo es realizado un estudio del funcionamiento de los mapas de kohonen en una y dos dimensiones y las características de este tipo de redes que trabajan de forma similar al cerebro humano. Finalmente, son detalladas las características necesarias para realizar el entrenamiento de las redes y la forma como son utilizados sus resultados, con la finalidad de descubrir características de la información de entrada, como por ejemplo, la distribución, la densidad y la forma de la informació
dc.publisherUniversidad Autónoma de Bucaramanga UNAB
dc.rightsDerechos de autor 2010 Revista Colombiana de Computación
dc.sourceRevista Colombiana de Computación; Vol. 11 Núm. 2 (2010): Revista Colombiana de Computación; 116-127
dc.titleRepresentación de información utilizando SOM (Mapas Autoorganizados) de Kohonen
dc.title.translatedRepresentation of information using Kohonen's SOM (Self-Organizing Maps)
dc.publisher.facultyFacultad Ingeniería
dc.publisher.programPregrado Ingeniería Mecatrónica
dc.subject.keywordsKohonen algorithmeng
dc.subject.keywordsUnsupervised neural networkseng
dc.subject.keywordsCompetitive trainingeng
dc.subject.keywordsMathematical logiceng
dc.identifier.instnameinstname:Universidad Autónoma de Bucaramanga UNABspa
dc.relation.referencesA. J. Smola, P. L. Bartlett, “Advances in neural information Processing systems”,MassachusettsInstituteofTechnology,2000
dc.relation.referencesT. Kohonen, Self-Organization andAssociative Memory, SpringerVerlag, BerlinHeildebergNewYork,1989.
dc.relation.referencesT. Kohonen, E. Oja, O. Simula, A. Visa, and J. Kangas, “Engineering applicationsoftheself-organizingmap,”inProc.IEEE,Oct.1996,vol.84, pp.1358–1384.
dc.relation.referencesGonzalezacuña,H.;DutraM.S;Lengerke,O..Identificationandmodeling for non-linear dynamic system using neural networks type MLP. In: Euro American Conference On Telematics And Information Systems, Prague, CzechRepublic.2009.v.26._Ref259054579
dc.relation.referencesS.V.N.Vishwanathan and M.Narasimha Murty. Use of kohonen map for dimensionality reduction. Technical Report TR No. IISC-CSA-1999-8, IndianInstituteofScience,ComputerScienceandAutomationDepartment, Bangalore,December1999.
dc.relation.referencesD. Michie, D.J. Spieegelhalter, C.C Taylor. “Machine learning, Neural and statisticalClasification”.February17,1994.
dc.relation.referencesR. M. Gray, “Vector quantization,” IEEE Acoust., Speech, Signal ProcessingMag.,pp.9–31,Apr.1984.
dc.relation.referencesS. Haykin, “Nueral Networks:Acomprehensive Foundation”. 2nd edition, PrenticeHall.
dc.relation.referencesSameer A. Nene and Shree K. Nayar. A simple algorithm for nearest neighbor search in high dimensions. IEEE transactions on Pattern Analysis andMachineIntelligence,19(9):989–1003,September1997.
dc.relation.referencesP.D.Wasserman. “Neural Computing theory and practice”, Vannostrand reinhold,NewYork1989.
dc.relation.referencesG. Burel, “A new approach for neural networks: The scalar distributed representation,” Traitement du Signal, vol. 10, no. 1, pp. 41–51, Apr. 1993.
dc.relation.referencesT. Kohonen, Speech Recognition Based on Topology-Preserving Neural Maps, in I. Aleksander (Ed.), Neural Computing Architectures, MIT Press,Cambridge,MA,1989,pp.26-40..
dc.contributor.cvlacGonzález Acuña, Hernán [0000774774]
dc.contributor.googlescholarGonzález Acuña, Hernán [NUgEExkAAAAJ]
dc.contributor.orcidGonzález Acuña, Hernán [0000-0003-2118-2272]
dc.contributor.scopusGonzález Acuña, Hernán [55942191000]
dc.contributor.scopusGonzález Acuña, Hernán [Hernan_Acuna2]
dc.subject.lembLógica matemáticaspa
dc.description.abstractenglishIn this paper is presented a demonstration of Kohonen's self-organizing maps,also known as SOM. Likewise is prepared a study of the functioning ofKohonen's maps in one and two dimensions and the most importantcharacteristics of this type of network that works in similar way that the humanbrain. Finally, this paper details the characteristics necessaries for thenetwork's training and how is possible use the results of the neural networks todiscover the characteristics of the information input for instance, how is your distribution, the density and shape.eng
dc.subject.proposalAlgoritmo de Kohonenspa
dc.subject.proposalRedes neuronales no supervisadasspa
dc.subject.proposalEntrenamiento competitivospa
dc.rights.creativecommonsAtribución-NoComercial-SinDerivadas 2.5 Colombia*
dc.contributor.researchgroupGrupo de Investigación Control y Mecatrónica - GICYMspa

Files in this item


This item appears in the following Collection(s)

Show simple item record

Atribución-NoComercial-SinDerivadas 2.5 Colombia
Except where otherwise noted, this item's license is described as Atribución-NoComercial-SinDerivadas 2.5 Colombia