{"id":"W4388439682","doi":"10.1139/cjfas-2022-0279","title":"Integrating disparate survey data in species distribution models demonstrate the need for robust model evaluation","year":2023,"lang":"en","type":"article","venue":"Canadian Journal of Fisheries and Aquatic Sciences","topic":"Species Distribution and Climate Change","field":"Environmental Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University; University of British Columbia; Fisheries and Oceans Canada","funders":"Natural Resources Canada; Fisheries and Oceans Canada; National Oceanic and Atmospheric Administration; Government of Canada","keywords":"Survey data collection; Sampling (signal processing); Computer science; Generalized linear mixed model; Geography; Statistics; Data mining; Machine learning; Mathematics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.07935474,0.001666345,0.001737374,0.002194609,0.001350485,0.003939145,0.003205933,0.002248721,0.0009167119],"category_scores_gemma":[0.1513253,0.001610639,0.002653816,0.002253368,0.002249363,0.004783825,0.004183163,0.003684956,0.000350402],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002178864,"about_ca_system_score_gemma":0.00235661,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02244032,"about_ca_topic_score_gemma":0.02709182,"domain_scores_codex":[0.9703076,0.02208667,0.001443126,0.00344964,0.002239489,0.00047356],"domain_scores_gemma":[0.8732862,0.1033747,0.006215162,0.01216007,0.004233289,0.0007304527],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001901181,0.000271722,0.1608834,0.0002552773,0.002334293,0.0001952308,0.0003901561,0.7664752,0.001370115,0.004997776,0.001049244,0.06158755],"study_design_scores_gemma":[0.00005566674,0.0002573084,0.02996529,0.0001478844,0.0002411667,0.0001058098,0.0003527736,0.9424287,0.001251336,0.02350557,0.001594771,0.00009378738],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2728239,0.0008722621,0.7202642,0.001475568,0.0001281771,0.0003765584,0.001059506,0.001044547,0.001955284],"genre_scores_gemma":[0.817503,0.0002044628,0.1794609,0.0005079926,0.00005971153,0.0003024048,0.001495583,0.0001741202,0.0002918359],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9206452,"threshold_uncertainty_score":0.419673,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2387536365564564,"score_gpt":0.3027070156722104,"score_spread":0.06395337911575397,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}