{"id":"W4407318277","doi":"10.1016/j.desal.2025.118678","title":"Robust deep learning model combined with missing input data estimation: Application in a 1000 m3/day high-salinity SWRO plant","year":2025,"lang":"en","type":"article","venue":"Desalination","topic":"Reservoir Engineering and Simulation Methods","field":"Engineering","cited_by":8,"is_retracted":false,"has_abstract":false,"ca_institutions":"PricewaterhouseCoopers (Canada)","funders":"Korea Environmental Industry and Technology Institute; Ministère de l’Éducation, Gouvernement de l’Ontario; National Research Foundation of Korea; Ministry of Environment","keywords":"Missing data; Salinity; Estimation; Environmental science; Engineering; Artificial intelligence; Computer science; Machine learning; Geology; Oceanography; Systems engineering","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":[0.0004937742,0.0005211328,0.0006161102,0.0002474225,0.0004827137,0.0004167022,0.0008529977,0.00121938,0.001259097],"category_scores_gemma":[0.001045844,0.0002979045,0.0004392413,0.000406013,0.000361211,0.0005953969,0.0005312491,0.0009066893,0.0001512492],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007498108,"about_ca_system_score_gemma":0.0008890929,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03684739,"about_ca_topic_score_gemma":0.03460758,"domain_scores_codex":[0.999864,0.00003022708,0.00000608765,0.00003569707,0.00004119525,0.00002275239],"domain_scores_gemma":[0.9995516,0.0002323118,0.00003497667,0.00004123779,0.0001032014,0.00003665135],"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":[0.0002514092,0.0002080467,0.00485751,0.00007361345,0.0000393885,0.0003155599,0.00005405558,0.9659601,0.005528567,0.0003664386,0.0008567006,0.02148858],"study_design_scores_gemma":[0.000007480889,0.00002629396,0.0009301127,0.000001033835,0.000003518576,0.000006610379,0.00001038092,0.9977337,0.001081186,0.0001300853,0.00006490685,0.000004728507],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9400092,0.0001428698,0.05617793,0.000400513,0.00003671431,0.00003247549,0.000346746,0.0009826607,0.001870874],"genre_scores_gemma":[0.9921514,0.00002173317,0.006857697,0.00001762873,0.000003662541,0.000008842584,0.0001155358,0.00001961497,0.0008039886],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03684739,"threshold_uncertainty_score":0.07326579,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03194591144272396,"score_gpt":0.2810436141504237,"score_spread":0.2490977027076998,"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."}}