Correo Electrónico
DNINFOA - SIA
Bibliotecas
Convocatorias
Identidad U.N.
Published
Representation and classification of geospatial data: a comparison of Kohonen's maps and the Neural Gas method
Representación y clasificación de datos geoespaciales: comparación entre mapas autoorganizativos de Kohonen y el método Gas Neuronal
DOI:
https://doi.org/10.15446/ing.investig.v23n3.14697Keywords:
Geographic information System (GIS), spatial data mining, neuronal networks, Self organizating map (SOM), kohonen, neural gas (en)Sistemas de Información geográfica (SIG), mineria de datos espaciales, redes neuronales, Mapas Autoorganizativos (SOM), Kohonen, Gas Neuronal (es)
Downloads
Approximately 80% of all the existing information in the world correspond to geo-referenced information. This creates an increasing necessity to have tools more flexible, precise and easy to use to the visualization, exploration and classification of great volumes of geospatial data. Additionally its necessary achieve smaller times to process this kind of information. In this preliminary investigation, different techniques are compared to visualize and to classify geo-referenced data using two types of neuronal networks: Kohonen's maps (SOM) and the Neural Gas method (NG). For the visualization cases, SOM showed a better performance than NG, occurring the opposite case for the classification examples.
Aproximadamente el 80% de toda la información existente en el mundo corresponde a información georreferenciada. Esto crea una creciente necesidad de disponer de herramientas más flexibles, precisas y fáciles de usar para la visualización, exploración y clasificación de grandes volúmenes de datos geoespaciales. En esta investigación preliminar se comparan diferentes técnicas para representar y clasificar datos georreferenciados utilizando dos tipos de redes neuronales: mapas auto organizativos de Kohonen (SOM) y el método Gas Neuronal (NG). El estudio incluye dos tipos de análisis: visualización y clasificación. Para el estudio correspondiente a visualización se escogieron dos tipos de datos: En primer lugar se seleccionó una muestra de 23000 coordenadas (x, y, z) de una zona montañosa de Colombia con el objetivo de analizar la capacidad de cada uno de los métodos para modelar el terreno, es decir, para presentar visualmente la forma del relieve. El segundo conjunto de datos corresponde a la población de cada uno de los 1090 municipios de Colombia (coordenadas x, y, y población total). El objetivo poder visualizar geográficamente la densidad poblacional de cada una de las regiones. Para el análisis de clasificación igualmente se seleccionaron dos conjuntos de datos: el primero corresponde a la codificación climática de los municipios de Colombia; el segundo, a la clasificación de los municipios en su respectivo departamento. Para los casos de visualización, SOM mostró un mejor desempeño que NG, dándose el caso contrario para los ejemplos de clasificación.
References
Kohonen, T. Self-Organizing Maps. Springer Series in Information Sciences, 30, Berlin 1995. DOI: https://doi.org/10.1007/978-3-642-97610-0
Martinetz, M.; Berkovich, S.; Schulten, K. Neural-gas network for vector quantization and its application to time series prediction, IEEE Trans. Neural Networks, V. 4, 1993, 558-569. DOI: https://doi.org/10.1109/72.238311
Bartkowiak, A.; Evelpidou, N.; Vassilopoulos, A. Choosing data vectors representing a huge data set: Kohonen's SOMs applied to the Kefallinia erosion data.
Vesanto, J.; Himberg, J.; Alhoniemi, E.; Parhankangas, I. SOM Joathar for Matlab 5. SOM Toolbox team, Helsinki University of Technology, Finland, Libella Oy, Espoo 2000, 1-54,
Ultsch, A. Self-organizing Neural Networks for Visualization and Classification. In O. Optiz; B. Lausen and R, Klar (Eds.). Information and Classification, Berlin: Springer-Verlag, 1999, 307-313. DOI: https://doi.org/10.1007/978-3-642-50974-2_31
Gahegan, M. On the application of inductive machine learning tools to geographical analysis. Geographical Analysis, Vol, 32, No. 2, 2000, 113-119, DOI: https://doi.org/10.1111/j.1538-4632.2000.tb00420.x
Heinke, D.; Hamker, F.H. Comparing neural network benchmarks on growing neural gas, growing cell structure, and fuzzy ARTMAP. IEEE Trans, on Neural Network, 1998, pp. 1279-1291. DOI: https://doi.org/10.1109/72.728377
How to Cite
APA
ACM
ACS
ABNT
Chicago
Harvard
IEEE
MLA
Turabian
Vancouver
Download Citation
CrossRef Cited-by
Dimensions
PlumX
Article abstract page views
Downloads
License
Copyright (c) 2003 Marly Esther De Moya Aarís
This work is licensed under a Creative Commons Attribution 4.0 International License.
The authors or holders of the copyright for each article hereby confer exclusive, limited and free authorization on the Universidad Nacional de Colombia's journal Ingeniería e Investigación concerning the aforementioned article which, once it has been evaluated and approved, will be submitted for publication, in line with the following items:
1. The version which has been corrected according to the evaluators' suggestions will be remitted and it will be made clear whether the aforementioned article is an unedited document regarding which the rights to be authorized are held and total responsibility will be assumed by the authors for the content of the work being submitted to Ingeniería e Investigación, the Universidad Nacional de Colombia and third-parties;
2. The authorization conferred on the journal will come into force from the date on which it is included in the respective volume and issue of Ingeniería e Investigación in the Open Journal Systems and on the journal's main page (https://revistas.unal.edu.co/index.php/ingeinv), as well as in different databases and indices in which the publication is indexed;
3. The authors authorize the Universidad Nacional de Colombia's journal Ingeniería e Investigación to publish the document in whatever required format (printed, digital, electronic or whatsoever known or yet to be discovered form) and authorize Ingeniería e Investigación to include the work in any indices and/or search engines deemed necessary for promoting its diffusion;
4. The authors accept that such authorization is given free of charge and they, therefore, waive any right to receive remuneration from the publication, distribution, public communication and any use whatsoever referred to in the terms of this authorization.
