{"id":"W3171394146","doi":"10.1016/j.envres.2021.111454","title":"A data-driven binary-classification framework for oil fingerprinting analysis","year":2021,"lang":"en","type":"article","venue":"Environmental Research","topic":"Oil Spill Detection and Mitigation","field":"Environmental Science","cited_by":31,"is_retracted":false,"has_abstract":false,"ca_institutions":"Memorial University of Newfoundland","funders":"Natural Sciences and Engineering Research Council of Canada; Pancreatic Cancer Action; Canada Research Chairs; Canada Foundation for Innovation","keywords":"Overfitting; Random forest; Classifier (UML); Principal component analysis; Computer science; Artificial intelligence; Binary classification; Support vector machine; Oil spill; Pattern recognition (psychology); Environmental science; Artificial neural network; Environmental engineering","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.0008648722,0.0001017891,0.0001277527,0.0001038364,0.0004625792,0.00008387944,0.0003923409,0.0001011916,0.006256778],"category_scores_gemma":[0.0004110371,0.0001086989,0.00009217638,0.000844844,0.0002393516,0.0002638122,0.0007553249,0.0002996945,0.001171943],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003752377,"about_ca_system_score_gemma":0.00001572809,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004527031,"about_ca_topic_score_gemma":0.0001330621,"domain_scores_codex":[0.9978235,0.000188311,0.0002104933,0.0006973502,0.0006981891,0.0003822019],"domain_scores_gemma":[0.9986559,0.0003165239,0.00005852655,0.0008287684,0.000007145506,0.0001331938],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0000546796,0.0006040999,0.1153565,0.00002776804,0.0002372469,0.00002018369,0.0005286395,0.003209797,0.3260944,0.0005517593,0.001563597,0.5517513],"study_design_scores_gemma":[0.0005761095,0.0001418452,0.6748614,0.00003113178,0.0001963412,0.000007535278,0.002284869,0.1859576,0.03018078,0.004298948,0.1009727,0.0004906694],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9812802,0.00008454687,0.01169282,0.001125705,0.00009828927,0.0001547636,0.000130071,0.00004125519,0.005392313],"genre_scores_gemma":[0.9777337,0.0001875725,0.0168712,0.00008815731,0.00007160607,0.00008496047,0.0005196636,0.00001892363,0.004424169],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5595049,"threshold_uncertainty_score":0.9996058,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1095088628269918,"score_gpt":0.378502824262011,"score_spread":0.2689939614350192,"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."}}