Discrimination of fish populations using parasites: Random Forests on a ‘predictable’ host-parasite system

0353458 - BC 2011 RIV GB eng J - Journal Article
Pérez-Del-Olmo, A. - Montero, E. E. - Fernández, M. - Barrett, J. - Raga, J. A. - Kostadinova, Aneta
Discrimination of fish populations using parasites: Random Forests on a ‘predictable’ host-parasite system.
Parasitology. Roč. 137, č. 12 (2010), s. 1833-1847. ISSN 0031-1820. E-ISSN 1469-8161
R&D Projects: GA MŠMT LC522
Institutional research plan: CEZ:AV0Z60220518
Keywords : predictive models * Random Forests * fish population discrimination * parasites as tags * Boops boops * Mediterranean * North-East Atlantic
Subject RIV: GJ - Animal Vermins ; Diseases, Veterinary Medicine
Impact factor: 2.522, year: 2010

We address the effect of spatial scale and temporal variation on model generality when forming predictive models for fish assignment using a new data mining approach, Random Forests (RF), to variable biological markers (parasite community data). Models were implemented for a fish host-parasite system sampled along the Mediterranean and Atlantic coasts of Spain. The main results are that (i) RF are well suited for multiclass population assignment using parasite communities in non-migratory fish; (ii) RF provide an efficient means for model cross-validation on the baseline data and this allows sample size limitations in parasite tag studies to be tackled effectively; (iii) the performance of RF is dependent on the complexity and spatial extent/configuration of the problem; and (iv) the development of predictive models is strongly influenced by seasonal change and this stresses the importance of both temporal replication and model validation in parasite tagging studies.
Permanent Link: http://hdl.handle.net/11104/0192704