Supervised pre-processing approaches in multiple class variables classification for fish recruitment forecasting

Handle URI:
http://hdl.handle.net/10754/562628
Title:
Supervised pre-processing approaches in multiple class variables classification for fish recruitment forecasting
Authors:
Fernandes, José Antonio; Lozano, Jose A.; Iñza, Iñaki; Irigoien, Xabier ( 0000-0002-5411-6741 ) ; Pérez, Aritz; Rodríguez, Juan Diego
Abstract:
A multi-species approach to fisheries management requires taking into account the interactions between species in order to improve recruitment forecasting of the fish species. Recent advances in Bayesian networks direct the learning of models with several interrelated variables to be forecasted simultaneously. These models are known as multi-dimensional Bayesian network classifiers (MDBNs). Pre-processing steps are critical for the posterior learning of the model in these kinds of domains. Therefore, in the present study, a set of 'state-of-the-art' uni-dimensional pre-processing methods, within the categories of missing data imputation, feature discretization and feature subset selection, are adapted to be used with MDBNs. A framework that includes the proposed multi-dimensional supervised pre-processing methods, coupled with a MDBN classifier, is tested with synthetic datasets and the real domain of fish recruitment forecasting. The correctly forecasting of three fish species (anchovy, sardine and hake) simultaneously is doubled (from 17.3% to 29.5%) using the multi-dimensional approach in comparison to mono-species models. The probability assessments also show high improvement reducing the average error (estimated by means of Brier score) from 0.35 to 0.27. Finally, these differences are superior to the forecasting of species by pairs. © 2012 Elsevier Ltd.
KAUST Department:
Biological and Environmental Sciences and Engineering (BESE) Division; Red Sea Research Center (RSRC); Marine Science Program; Plankton ecology Research Group
Publisher:
Elsevier
Journal:
Environmental Modelling and Software
Issue Date:
Feb-2013
DOI:
10.1016/j.envsoft.2012.10.001
Type:
Article
ISSN:
13648152
Sponsors:
Jose A. Fernandes is supported by a Doctoral Fellowship from the Fundacion Centros Tecnologicos Inaki Goenaga. This work has been supported, partially, by the Etortek, Saiotek and Research Groups 2007-2012 (IT-242-07) programmes (Basque Government), TIN2010-14931 and Consolider Ingenio 2010-CSD2007-00018 projects (Spanish Ministry of Education and Science) and COMBIOMED network in computational biomedicine (Carlos III Health Institute). This research is funded partially by the project ECOANCHOA, funded by the Department of Agriculture, Fisheries and Food of the Basque Country Government and the VII Framework projects MEECE No 212085 and FACTS no 244966. This is contribution 593 from the Marine Research Division (AZTI-Tecnalia).
Appears in Collections:
Articles; Red Sea Research Center (RSRC); Marine Science Program; Biological and Environmental Sciences and Engineering (BESE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorFernandes, José Antonioen
dc.contributor.authorLozano, Jose A.en
dc.contributor.authorIñza, Iñakien
dc.contributor.authorIrigoien, Xabieren
dc.contributor.authorPérez, Aritzen
dc.contributor.authorRodríguez, Juan Diegoen
dc.date.accessioned2015-08-03T10:58:57Zen
dc.date.available2015-08-03T10:58:57Zen
dc.date.issued2013-02en
dc.identifier.issn13648152en
dc.identifier.doi10.1016/j.envsoft.2012.10.001en
dc.identifier.urihttp://hdl.handle.net/10754/562628en
dc.description.abstractA multi-species approach to fisheries management requires taking into account the interactions between species in order to improve recruitment forecasting of the fish species. Recent advances in Bayesian networks direct the learning of models with several interrelated variables to be forecasted simultaneously. These models are known as multi-dimensional Bayesian network classifiers (MDBNs). Pre-processing steps are critical for the posterior learning of the model in these kinds of domains. Therefore, in the present study, a set of 'state-of-the-art' uni-dimensional pre-processing methods, within the categories of missing data imputation, feature discretization and feature subset selection, are adapted to be used with MDBNs. A framework that includes the proposed multi-dimensional supervised pre-processing methods, coupled with a MDBN classifier, is tested with synthetic datasets and the real domain of fish recruitment forecasting. The correctly forecasting of three fish species (anchovy, sardine and hake) simultaneously is doubled (from 17.3% to 29.5%) using the multi-dimensional approach in comparison to mono-species models. The probability assessments also show high improvement reducing the average error (estimated by means of Brier score) from 0.35 to 0.27. Finally, these differences are superior to the forecasting of species by pairs. © 2012 Elsevier Ltd.en
dc.description.sponsorshipJose A. Fernandes is supported by a Doctoral Fellowship from the Fundacion Centros Tecnologicos Inaki Goenaga. This work has been supported, partially, by the Etortek, Saiotek and Research Groups 2007-2012 (IT-242-07) programmes (Basque Government), TIN2010-14931 and Consolider Ingenio 2010-CSD2007-00018 projects (Spanish Ministry of Education and Science) and COMBIOMED network in computational biomedicine (Carlos III Health Institute). This research is funded partially by the project ECOANCHOA, funded by the Department of Agriculture, Fisheries and Food of the Basque Country Government and the VII Framework projects MEECE No 212085 and FACTS no 244966. This is contribution 593 from the Marine Research Division (AZTI-Tecnalia).en
dc.publisherElsevieren
dc.subjectBayesian networksen
dc.subjectDiscretizationen
dc.subjectEnvironmental modellingen
dc.subjectFeature subset selectionen
dc.subjectMissing imputationen
dc.subjectMulti-dimensional classificationen
dc.subjectRecruitment forecastingen
dc.subjectSupervised classificationen
dc.titleSupervised pre-processing approaches in multiple class variables classification for fish recruitment forecastingen
dc.typeArticleen
dc.contributor.departmentBiological and Environmental Sciences and Engineering (BESE) Divisionen
dc.contributor.departmentRed Sea Research Center (RSRC)en
dc.contributor.departmentMarine Science Programen
dc.contributor.departmentPlankton ecology Research Groupen
dc.identifier.journalEnvironmental Modelling and Softwareen
dc.contributor.institutionAZTI-Tecnalia, Marine Research Division, Herrera Kaia z/g, E-20110 Pasaia (Gipuzkoa), Spainen
dc.contributor.institutionUniversity of the Basque Country, Department of Computer Science and AI, Intelligent Systems Group (ISG), Paseo Manuel de Lardizabal, 1, E-20018 Donostia - San Sebastián, Spainen
kaust.authorIrigoien, Xabieren
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.