@article{fdi:010081073,
  title = {{M}achine learning to detect bycatch risk : novel application to echosounder buoys data in tuna purse seine fisheries},
  author = {{M}annocci, {L}. and {B}aidai, {Y}. and {F}orget, {F}abien and {T}olotti, {M}. {T}. and {D}agorn, {L}aurent and {C}apello, {M}anuela},
  editor = {},
  language = {{ENG}},
  abstract = {{T}he advent of big data and machine learning offers great promise for addressing conservation and management questions in the oceans. {Y}et, few applications of machine learning exist to mitigate the overexploitation of marine resources. {T}ropical tuna purse seine fisheries ({TTPSF}) are distributed worldwide and account for two thirds of the global tuna catch. {I}n these fisheries, the use of {D}rifting {F}ish {A}ggregating {D}evices ({DFAD}s)? n-made floating objects massively deployed by fishers to increase their tuna catches?results in the incidental catch of non-target species, termed bycatch. {W}e explored the possibility of applying machine learning on echosounder buoys attached to {DFAD}s, representing an unprecedented source of big data, for identifying high bycatch risk at {DFAD}s. {W}e trained random forests algorithms to differentiate between high and low bycatch occurrence based on matched echosounder and onboard observer data for the same {DFAD}s (representing sample sizes of 838 and 2144 in the {A}tlantic and the {I}ndian {O}cean, respectively). {A}lgorithms showed a better performance in the {A}tlantic {O}cean (accuracy of 0.66 versus 0.58 in the {I}ndian {O}cean) and were best at detecting the ?high bycatch? occurrence class. {T}hese results unravel the potential of machine learning applied to fishers? buoys data for bycatch reduction and improved selectivity in one of the largest fisheries worldwide.},
  keywords = {{A}tlantic {O}cean ; {D}rifting fish aggregating devices ; {E}chosounder buoys ; {I}ndian {O}cean ; {R}andom forests ; {T}ropical tuna purse seine fisheries ; {ATLANTIQUE}},
  booktitle = {},
  journal = {{B}iological {C}onservation},
  volume = {255},
  numero = {},
  pages = {109004 [6 ]},
  ISSN = {0006-3207},
  year = {2021},
  DOI = {10.1016/j.biocon.2021.109004},
  URL = {https://www.documentation.ird.fr/hor/fdi:010081073},
}