{"id":"W4413092155","doi":"10.2139/ssrn.5354381","title":"Prediction of Ambient Pm2.5 Chemical Components in Southern California Using Machine Learning","year":2025,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Air Quality Monitoring and Forecasting","field":"Environmental Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Environmental science; Artificial intelligence; Machine learning; Computer science","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":"codex-gemma-dda1882f352a","candidate_categories":["research_integrity"],"consensus_categories":[],"category_scores_codex":[0.002030932,0.000229266,0.0003369856,0.0001160604,0.0001336708,0.00002415715,0.0003344535,0.0002267989,0.00005052707],"category_scores_gemma":[0.00009258735,0.0002259814,0.0001511089,0.0001678627,0.00008983081,0.00004488788,0.0005869637,0.004674608,0.00001763863],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003033847,"about_ca_system_score_gemma":0.0002982264,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004872533,"about_ca_topic_score_gemma":0.0002896718,"domain_scores_codex":[0.9971513,0.0002137991,0.0006335006,0.0003345832,0.0004596139,0.001207248],"domain_scores_gemma":[0.9992558,0.00003859086,0.0004555099,0.0001647014,0.00001607286,0.00006927169],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001442946,0.0002713063,0.6927667,0.0001032253,0.0001486975,0.000005376052,0.001613425,0.2578147,0.03658024,0.00006765944,0.00000517612,0.01047914],"study_design_scores_gemma":[0.003860554,0.0004887586,0.01374847,0.003357222,0.0004435188,0.0004621847,0.006749121,0.9026291,0.00835231,0.05730664,0.001032459,0.001569685],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9930919,0.0003951671,0.005825762,0.0000456567,0.0002664263,0.0001274184,0.00007755141,0.00002664191,0.0001435166],"genre_scores_gemma":[0.9987097,0.0002079815,0.0004666326,0.000005936176,0.000152158,0.000003269563,0.00003857977,0.00001872832,0.0003970061],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6790183,"threshold_uncertainty_score":0.9976217,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03472033055528631,"score_gpt":0.2531452383006919,"score_spread":0.2184249077454055,"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."}}