@article{fdi:010053066,
  title = {{O}ptimization of an artificial neural network for identifying fishing set positions from {VMS} data : an example from the {P}eruvian anchovy purse seine fishery},
  author = {{J}oo, {R}. and {B}ertrand, {S}ophie and {C}haigneau, {A}lexis and {N}iquen, {M}.},
  editor = {},
  language = {{ENG}},
  abstract = {{T}he spatial behavior of numerous fishing fleets is nowadays well documented thanks to satellite {V}essel {M}onitoring {S}ystems ({VMS}). {V}essel positions are recorded on a frequent and regular basis which opens promising perspectives for improving fishing effort estimation and management. {H}owever, no specific information is provided on whether the vessel is fishing or not. {T}o answer that question, existing works on {VMS} data usually apply simple criteria (e.g. threshold on speed). {T}hose simple criteria generally focus in detecting true positives (a true fishing set detected as a fishing set); conversely, estimation errors are given no attention. {F}or our case study, the {P}eruvian anchovy fishery, those criteria overestimate the total number of fishing sets by 182%. {T}o overcome this problem an artificial neural network ({ANN}) approach is presented here. {I}n order to set both the optimal parameterization and use "rules" for this {ANN}, we perform an extensive sensitivity analysis on the optimization of (1) the internal structure and training algorithm of the {ANN} and (2) the "rules" used for choosing both the relative size and the composition of the databases ({DB}s) used for training and inferring with the {ANN}. {T}he "optimized" {ANN} greatly improves the estimates of the number and location of fishing events. {F}or our case study, {ANN} reduces the total estimation error on the number of fishing sets to 1% (in average) and obtains 76% of true positives. {T}his spatially explicit information on effort, provided with error estimation, should greatly reduce misleading interpretations of catch per unit effort and thus significantly improve the adaptive management of fisheries. {W}hile fitted on {P}eruvian anchovy fishery data, this type of neural network approach has wider potential and could be implemented in any fishery relying on both {VMS} and at-sea observer data. {I}n order to increase the accuracy of the {ANN} results, we also suggest some criteria for improving sampling design by at-sea observers and {VMS} data.},
  keywords = {{A}rtificial neural networks ; {S}ensitivity analysis ; {F}ishing set ; locations ; {V}essel {M}onitoring {S}ystem ; {A}nchovy purse seine fishery},
  booktitle = {},
  journal = {{E}cological {M}odelling},
  volume = {222},
  numero = {4},
  pages = {1048--1059},
  ISSN = {0304-3800},
  year = {2011},
  DOI = {10.1016/j.ecolmodel.2010.08.039},
  URL = {https://www.documentation.ird.fr/hor/fdi:010053066},
}