{"id":"W4292836181","doi":"10.1111/1365-2664.14280","title":"<i>NABat ML</i>: Utilizing deep learning to enable crowdsourced development of automated, scalable solutions for documenting North American bat populations","year":2022,"lang":"en","type":"article","venue":"Journal of Applied Ecology","topic":"Bat Biology and Ecology Studies","field":"Agricultural and Biological Sciences","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"Parks Canada; Government of Newfoundland and Labrador; University of Prince Edward Island","funders":"U.S. Geological Survey","keywords":"Computer science; Scalability; Artificial intelligence; Convolutional neural network; Machine learning; Robustness (evolution); Bat algorithm; Implementation; Citizen science; Deep learning; Cloud computing; Human echolocation; Database; Software engineering","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001237594,0.0007764827,0.0002877794,0.0007220299,0.0004321494,0.001192087,0.001723535,0.0008829795,0.003499683],"category_scores_gemma":[0.002749632,0.0002766441,0.0003920063,0.0004245085,0.0005321768,0.001746074,0.001528736,0.001034807,0.00161966],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009582101,"about_ca_system_score_gemma":0.0009427417,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02570203,"about_ca_topic_score_gemma":0.04166632,"domain_scores_codex":[0.9995363,0.00009667587,0.00002358198,0.0001464954,0.0001458287,0.00005108506],"domain_scores_gemma":[0.9987566,0.0002538439,0.0000857851,0.0002940232,0.0005083986,0.0001012734],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003200016,0.0005131954,0.01195575,0.0001792643,0.0002274238,0.0003421314,0.0002719766,0.1923315,0.0464161,0.005684113,0.1974429,0.5443155],"study_design_scores_gemma":[0.00001363104,0.00004301793,0.001392291,0.000009502399,0.000009478514,0.00001725956,0.00003678613,0.9770536,0.01099356,0.001608894,0.008805725,0.00001620324],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1194857,0.000557569,0.7807291,0.006519347,0.001101073,0.0005533196,0.005519039,0.06034528,0.02518954],"genre_scores_gemma":[0.5190335,0.0002853358,0.4445658,0.001920889,0.0003212207,0.00038352,0.01212552,0.00151125,0.01985291],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02570203,"threshold_uncertainty_score":0.05110484,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02815345821664952,"score_gpt":0.2465415904317624,"score_spread":0.2183881322151128,"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."}}