@article{fdi:010020707,
  title = {{A}rtificial neural networks as a tool in ecological modelling : an introduction},
  author = {{L}ek, {S}. and {G}u{\'e}gan, {J}ean-{F}ran{\c{c}}ois},
  editor = {},
  language = {{ENG}},
  abstract = {{A}rtificial neural networks ({ANN}s) are non-linear mapping structures based on the function of the human brain. {T}hey have been shown to be universal and highly flexible function approximators for any data. {T}hese make powerful tools for models, especially when the underlying data relationships are unknown. {F}or this reason, the international workshop on the applications of {ANN}s to ecological modelling was organized in {T}oulouse, {F}rance ({D}ecember 1998). {D}uring this meeting, we discussed different methods, and their reliability to deal with ecological data. {T}he special issue of this ecological modelling journal begins with the state-of-the-art with emphasis on the development of structural dynamic models presented by {S}.{E}. {J}orgensen ({DK}). {T}hen, to illustrate the ecological applications of {ANN}s, examples are drawn from several fields, e.g. terrestrial and aquatic ecosystems, remote sensing and evolutionary ecology. {I}n this paper, we present some of the most important papers of the first workshop about {ANN}s in ecological modelling. {W}e briefly introduce here two algorithms frequently used ; (i) one supervised network, the backpropagation algorithm ; and (ii) one unsupervised network, the {K}ohonen self-organizing mapping algorithm. {T}he future development of {ANN}s is discussed in the present work. {S}everal examples of modelling of {ANN}s in various areas of ecology are presented in this special issue. ({R}{\'e}sum{\'e} d'auteur)},
  keywords = {{INTELLIGENCE} {ARTIFICIELLE} ; {MODELISATION} ; {ECOLOGIE} ; {METHODE} {D}'{ANALYSE} ; {ALGORITHME} ; {PREVISION}},
  booktitle = {{A}pplication of artificial neural networks in ecological modelling},
  journal = {{E}cological {M}odelling},
  volume = {120},
  numero = {2-3},
  pages = {65--73},
  ISSN = {0304-3800},
  year = {1999},
  DOI = {10.1016/{S}0304-3800(99)00092-7},
  URL = {https://www.documentation.ird.fr/hor/fdi:010020707},
}