{"id":"W4313253662","doi":"10.1016/j.jhazmat.2022.130633","title":"Appraisal of machine learning techniques for predicting emerging disinfection byproducts in small water distribution networks","year":2022,"lang":"en","type":"article","venue":"Journal of Hazardous Materials","topic":"Water Treatment and Disinfection","field":"Environmental Science","cited_by":36,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université Laval; Okanagan University College; University of British Columbia, Okanagan Campus; University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada; Université Laval","keywords":"Machine learning; Artificial neural network; Support vector machine; Multivariate adaptive regression splines; Bayesian multivariate linear regression; Kriging; Regression analysis; Chloropicrin; Artificial intelligence; Linear regression; Regression; Nonparametric regression; Computer science; Multivariate statistics; Statistics; Mathematics","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":[],"consensus_categories":[],"category_scores_codex":[0.001313534,0.0001075248,0.0002361976,0.00007071438,0.0001800014,0.00003123623,0.00008295863,0.00003535282,0.0003110104],"category_scores_gemma":[0.00007069309,0.00008082678,0.0000716747,0.0001023046,0.00002576266,0.0002036524,0.0001134249,0.0001197707,0.000001146389],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002161365,"about_ca_system_score_gemma":0.000005801192,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002877179,"about_ca_topic_score_gemma":0.00002501963,"domain_scores_codex":[0.9988149,0.0001752397,0.0005155756,0.0001348922,0.0001589614,0.0002004113],"domain_scores_gemma":[0.9994531,0.00003948668,0.0003906795,0.00006809994,0.00001956235,0.0000290393],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0006317892,0.0003021688,0.1996207,0.0000389608,0.00003372431,0.0000113104,0.0005495494,0.04972872,0.7445889,0.00001057284,0.0001455445,0.004338171],"study_design_scores_gemma":[0.001042783,0.001245456,0.02861657,0.00004393069,0.00008048405,0.00012288,0.0001266852,0.00144241,0.9633271,0.0003020075,0.003471989,0.0001777339],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9935136,0.00002921682,0.005293938,0.0001001135,0.0007473651,0.0002326926,0.00001917969,0.00002376315,0.00004015344],"genre_scores_gemma":[0.9992912,0.00001364403,0.0003416562,0.000005200881,0.000159988,0.00002635012,0.0001091732,0.00001436809,0.00003837295],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2187382,"threshold_uncertainty_score":0.3405345,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00799650678384332,"score_gpt":0.2315582798540563,"score_spread":0.223561773070213,"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."}}