{"id":"W2790034089","doi":"10.1016/j.watres.2018.02.052","title":"Neural networks for dimensionality reduction of fluorescence spectra and prediction of drinking water disinfection by-products","year":2018,"lang":"en","type":"article","venue":"Water Research","topic":"Water Quality Monitoring and Analysis","field":"Environmental Science","cited_by":94,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo; University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; University of Toronto; Canadian Water Network","keywords":"Interpretability; Overfitting; Autoencoder; Principal component analysis; Artificial neural network; Dimensionality reduction; Artificial intelligence; Biological system; Pattern recognition (psychology); Curse of dimensionality; Computer science; Representation (politics); Machine learning; Chemistry","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":[0.0008231492,0.0006352481,0.0005595517,0.0004995824,0.0003901327,0.0005797443,0.000581072,0.0006663415,0.0009309217],"category_scores_gemma":[0.002257526,0.0003148111,0.0008236208,0.0005201391,0.0002621566,0.0006733632,0.0005322164,0.001117086,0.0002351235],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008722716,"about_ca_system_score_gemma":0.0006491854,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01229878,"about_ca_topic_score_gemma":0.009690142,"domain_scores_codex":[0.9997329,0.00008770916,0.00002115662,0.0000687065,0.00004984537,0.00003965974],"domain_scores_gemma":[0.999335,0.0004139861,0.00005193279,0.0000495095,0.0001346341,0.00001498834],"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":[0.000212697,0.000181464,0.002238437,0.0000720648,0.0001350762,0.00003328366,0.00005641751,0.7119665,0.007135818,0.002299049,0.001970656,0.2736985],"study_design_scores_gemma":[0.000001804381,0.000007225287,0.0003214885,0.00000165659,0.0000047041,0.000002452424,0.000002773694,0.9983701,0.0006609492,0.0005504499,0.00007372886,0.000002777061],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2678833,0.002236929,0.7252564,0.0007067492,0.000141568,0.00008275265,0.0006162309,0.001082494,0.001993594],"genre_scores_gemma":[0.8619063,0.0005929908,0.1318258,0.00009858242,0.00007633342,0.000150149,0.0009554942,0.0000522964,0.004342],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01229878,"threshold_uncertainty_score":0.02445441,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05297307805395812,"score_gpt":0.315411533021768,"score_spread":0.2624384549678099,"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."}}