{"id":"W4312971753","doi":"10.1039/d2ea00084a","title":"Application of machine learning and statistical modeling to identify sources of air pollutant levels in Kitchener, Ontario, Canada","year":2022,"lang":"en","type":"article","venue":"Environmental Science Atmospheres","topic":"Air Quality and Health Impacts","field":"Environmental Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; Continental (Canada); Wilfrid Laurier University","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Foundation for Innovation","keywords":"Pollutant; Environmental science; Air pollutants; Meteorology; Air pollution; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007630847,0.0003952456,0.0003342668,0.001183807,0.002598102,0.001180538,0.0009432292,0.0003566963,0.00211442],"category_scores_gemma":[0.002527422,0.0003771333,0.000584697,0.001889281,0.0004464409,0.0003059889,0.0006154716,0.0006026108,0.0002728686],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.03156365,"about_ca_system_score_gemma":0.05161488,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.9982123,"about_ca_topic_score_gemma":0.9989147,"domain_scores_codex":[0.9993531,0.0001392755,0.00003965035,0.00009217012,0.000255,0.0001208422],"domain_scores_gemma":[0.99867,0.0002803436,0.0001016632,0.00004008748,0.0008067915,0.0001010928],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0004406493,0.000307958,0.7376452,0.0004243369,0.0004854593,0.000795681,0.003224514,0.1068275,0.002959292,0.00590001,0.01962829,0.1213611],"study_design_scores_gemma":[0.00006634028,0.00007564991,0.7778929,0.0001098959,0.0001276113,0.00008505565,0.003579778,0.1948274,0.001754732,0.001616537,0.01978595,0.00007811675],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9404896,0.002105779,0.02029081,0.003035621,0.00009982152,0.000304233,0.008713722,0.0003240882,0.02463644],"genre_scores_gemma":[0.9743356,0.0008183795,0.00898103,0.00009019727,0.00001118136,0.00006096816,0.002410156,0.0000464124,0.01324619],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03156365,"threshold_uncertainty_score":0.2290114,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01647046729034847,"score_gpt":0.274190913781408,"score_spread":0.2577204464910596,"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."}}