{"id":"W2952113774","doi":"10.1111/2041-210x.13120","title":"Machine learning to classify animal species in camera trap images: Applications in ecology","year":2018,"lang":"en","type":"article","venue":"Methods in Ecology and Evolution","topic":"Wildlife Ecology and Conservation","field":"Environmental Science","cited_by":481,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Saskatchewan","funders":"National Energy Technology Laboratory; Animal and Plant Health Inspection Service; University of Saskatchewan; National Wildlife Research Center; Colorado Parks and Wildlife; University of Georgia Research Foundation; University of Wyoming; U.S. Department of Agriculture; U.S. Department of Energy","keywords":"Artificial intelligence; Computer science; Camera trap; Wildlife; Laptop; Convolutional neural network; Sample (material); Citizen science; Machine learning; Cartography; Geography; Ecology","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001524957,0.000109853,0.0002037058,0.0002151706,0.0001615631,0.000005908959,0.0001148073,0.0002460696,0.0004867258],"category_scores_gemma":[0.0003529783,0.0001168396,0.00001680233,0.0005346839,0.0004122688,0.0001481574,0.0001179379,0.0003531541,0.00009754496],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003826841,"about_ca_system_score_gemma":0.00002131965,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005083311,"about_ca_topic_score_gemma":0.03765652,"domain_scores_codex":[0.9981486,0.0008313843,0.0002926782,0.000362604,0.00004312887,0.0003216341],"domain_scores_gemma":[0.9993519,0.0004236195,0.00006713697,0.0001051633,0.000008017515,0.0000441755],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00006503553,0.00009101834,0.9926175,0.000002754527,0.00000214851,0.000002513806,0.0005476518,0.0004028531,0.002759968,0.0006895324,0.0001014535,0.002717501],"study_design_scores_gemma":[0.0003818741,0.0002758933,0.9878868,0.000003260946,0.000004395285,0.000006431886,0.000344767,0.004488466,0.000114391,0.003878055,0.002503425,0.0001122747],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9863719,0.0000378191,0.00872125,0.00173918,0.0001240751,0.0003359416,0.000001679757,0.00001965066,0.002648497],"genre_scores_gemma":[0.9613001,0.00002904281,0.03755362,0.0004796899,0.0000394307,0.0001832305,0.000006241766,0.000006300225,0.0004023187],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03714819,"threshold_uncertainty_score":0.9799038,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0209818672774687,"score_gpt":0.3171276760884255,"score_spread":0.2961458088109568,"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."}}