Application of multiple self-organizing maps for classification of soil samples in Thailand according to their geographic origins

Copyright © 2016 John Wiley & Sons, Ltd. Multiple self-organizing maps (SOMs) were applied to classify soil samples according to their geographic origins. The soil physical and chemical parameters, including textures, pH, and chemical nutrients, were analyzed and used for establishing the chem...

Full description

Saved in:
Bibliographic Details
Main Authors: Chanida Krongchai, Sujitra Funsueb, Jaroon Jakmunee, Sila Kittiwachana
Format: Journal
Published: 2018
Subjects:
Online Access:https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85012898816&origin=inward
http://cmuir.cmu.ac.th/jspui/handle/6653943832/46603
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Chiang Mai University
id th-cmuir.6653943832-46603
record_format dspace
spelling th-cmuir.6653943832-466032018-04-25T07:27:42Z Application of multiple self-organizing maps for classification of soil samples in Thailand according to their geographic origins Chanida Krongchai Sujitra Funsueb Jaroon Jakmunee Sila Kittiwachana Mathematics Agricultural and Biological Sciences Copyright © 2016 John Wiley & Sons, Ltd. Multiple self-organizing maps (SOMs) were applied to classify soil samples according to their geographic origins. The soil physical and chemical parameters, including textures, pH, and chemical nutrients, were analyzed and used for establishing the chemometric models. To determine the optimum size and arrangement of the maps, we adapted a growing self-organizing map algorithm. To evaluate the reliability of the models, we calculated statistic indices based on the majority vote including percentage predictive ability, percentage model stability, and percentage correctly classified using a bootstrap methodology. For means of comparison, we also used linear discriminant analysis, quadratic discriminant analysis, partial least squares-discriminant analysis, soft independent modeling of class analogy, counter propagation network, supervised Kohonen network, and k-nearest neighbors. In comparison to a single SOM, multiple SOMs clearly provided better classification results. The extension of multiple SOMs also led to the best discrimination of the soil origins. 2018-04-25T06:57:51Z 2018-04-25T06:57:51Z 2017-02-01 Journal 1099128X 08869383 2-s2.0-85012898816 10.1002/cem.2871 https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85012898816&origin=inward http://cmuir.cmu.ac.th/jspui/handle/6653943832/46603
institution Chiang Mai University
building Chiang Mai University Library
country Thailand
collection CMU Intellectual Repository
topic Mathematics
Agricultural and Biological Sciences
spellingShingle Mathematics
Agricultural and Biological Sciences
Chanida Krongchai
Sujitra Funsueb
Jaroon Jakmunee
Sila Kittiwachana
Application of multiple self-organizing maps for classification of soil samples in Thailand according to their geographic origins
description Copyright © 2016 John Wiley & Sons, Ltd. Multiple self-organizing maps (SOMs) were applied to classify soil samples according to their geographic origins. The soil physical and chemical parameters, including textures, pH, and chemical nutrients, were analyzed and used for establishing the chemometric models. To determine the optimum size and arrangement of the maps, we adapted a growing self-organizing map algorithm. To evaluate the reliability of the models, we calculated statistic indices based on the majority vote including percentage predictive ability, percentage model stability, and percentage correctly classified using a bootstrap methodology. For means of comparison, we also used linear discriminant analysis, quadratic discriminant analysis, partial least squares-discriminant analysis, soft independent modeling of class analogy, counter propagation network, supervised Kohonen network, and k-nearest neighbors. In comparison to a single SOM, multiple SOMs clearly provided better classification results. The extension of multiple SOMs also led to the best discrimination of the soil origins.
format Journal
author Chanida Krongchai
Sujitra Funsueb
Jaroon Jakmunee
Sila Kittiwachana
author_facet Chanida Krongchai
Sujitra Funsueb
Jaroon Jakmunee
Sila Kittiwachana
author_sort Chanida Krongchai
title Application of multiple self-organizing maps for classification of soil samples in Thailand according to their geographic origins
title_short Application of multiple self-organizing maps for classification of soil samples in Thailand according to their geographic origins
title_full Application of multiple self-organizing maps for classification of soil samples in Thailand according to their geographic origins
title_fullStr Application of multiple self-organizing maps for classification of soil samples in Thailand according to their geographic origins
title_full_unstemmed Application of multiple self-organizing maps for classification of soil samples in Thailand according to their geographic origins
title_sort application of multiple self-organizing maps for classification of soil samples in thailand according to their geographic origins
publishDate 2018
url https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85012898816&origin=inward
http://cmuir.cmu.ac.th/jspui/handle/6653943832/46603
_version_ 1681422905191170048