TY - JOUR
T1 - Supervised pre-processing approaches in multiple class variables classification for fish recruitment forecasting
AU - Fernandes, José Antonio
AU - Lozano, Jose A.
AU - Iñza, Iñaki
AU - Irigoien, Xabier
AU - Pérez, Aritz
AU - Rodríguez, Juan Diego
N1 - KAUST Repository Item: Exported on 2020-10-01
Acknowledgements: 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).
PY - 2013/2
Y1 - 2013/2
N2 - 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.
AB - 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.
UR - http://hdl.handle.net/10754/562628
UR - https://linkinghub.elsevier.com/retrieve/pii/S1364815212002472
UR - http://www.scopus.com/inward/record.url?scp=84871748990&partnerID=8YFLogxK
U2 - 10.1016/j.envsoft.2012.10.001
DO - 10.1016/j.envsoft.2012.10.001
M3 - Article
SN - 1364-8152
VL - 40
SP - 245
EP - 254
JO - Environmental Modelling and Software
JF - Environmental Modelling and Software
ER -