{"id":"W2952297896","doi":"10.1016/j.envint.2019.104934","title":"A comparison of linear regression, regularization, and machine learning algorithms to develop Europe-wide spatial models of fine particles and nitrogen dioxide","year":2019,"lang":"en","type":"article","venue":"Environment International","topic":"Air Quality and Health Impacts","field":"Environmental Science","cited_by":338,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"China Scholarship Council; Medical Research Council; Health Effects Institute; U.S. Environmental Protection Agency","keywords":"Linear regression; Algorithm; Support vector machine; Random forest; Artificial neural network; Stepwise regression; Machine learning; Mathematics; Regression analysis; Statistics; Linear model; Artificial intelligence; 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":[],"consensus_categories":[],"category_scores_codex":[0.0003131482,0.00009684593,0.0001723527,0.00003585363,0.00005338011,0.000006242197,0.000107556,0.00003707592,0.0006126227],"category_scores_gemma":[0.000145779,0.00008592488,0.00001231543,0.00007462873,0.00009243724,0.0001417242,0.00027564,0.00008669579,0.00004259646],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004656716,"about_ca_system_score_gemma":0.000009721369,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000194478,"about_ca_topic_score_gemma":0.00002275339,"domain_scores_codex":[0.9988531,0.00007291958,0.0003607259,0.0002099118,0.0003670552,0.0001362344],"domain_scores_gemma":[0.9994981,0.00007898774,0.0001893747,0.0001035428,0.00001372369,0.0001162761],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00008225709,0.0001119417,0.9284573,0.00001861784,0.00001892151,6.488618e-7,0.0017883,0.04785423,0.00673804,0.0002908688,0.0001623318,0.01447651],"study_design_scores_gemma":[0.0008505424,0.0003273394,0.4791289,0.00008366109,0.00001606068,0.000004265843,0.0001327836,0.4640355,0.0405959,0.001347219,0.01326593,0.0002118859],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9388983,0.00007501203,0.0586815,0.001793872,0.00003764451,0.00018479,0.00002333308,0.000008383283,0.000297177],"genre_scores_gemma":[0.9734067,0.00007614593,0.02570732,0.000225922,0.00001679774,0.000004463908,0.0000427031,0.00001076285,0.0005091316],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4493285,"threshold_uncertainty_score":0.6707789,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0412298700638622,"score_gpt":0.2957456222841378,"score_spread":0.2545157522202756,"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."}}