{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002980861,0.0007613777,0.0006365063,0.001488248,0.0002954296,0.001121106,0.0007039631,0.0007473695,0.0006590852],"category_scores_gemma":[0.008613366,0.0001611756,0.0005761006,0.0009179718,0.0003200693,0.0009377952,0.000218982,0.000588062,0.000193441],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001057019,"about_ca_system_score_gemma":0.0005796678,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01023156,"about_ca_topic_score_gemma":0.007071974,"domain_scores_codex":[0.9989767,0.0004640002,0.00006259752,0.0001083837,0.0003486967,0.00003961788],"domain_scores_gemma":[0.988876,0.008692701,0.0003727323,0.0002012015,0.001751615,0.0001056018],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000983668,0.0004172507,0.03296705,0.0006089884,0.0003609582,0.000199997,0.00005567748,0.6050678,0.005775797,0.002313937,0.00495503,0.3462938],"study_design_scores_gemma":[0.0000173121,0.0002346617,0.006255646,0.00003244914,0.00004307731,0.000025308,0.00003417583,0.9902998,0.001620861,0.00084723,0.0005811711,0.000008270914],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7808781,0.02465006,0.1790611,0.0031241,0.0005614455,0.0002767542,0.0008088437,0.0009038301,0.009735813],"genre_scores_gemma":[0.9577022,0.004083373,0.03579322,0.000103395,0.0002074019,0.00004075045,0.0004667042,0.00002589861,0.001577131],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01023156,"threshold_uncertainty_score":0.02034402,"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."}}