{"id":"W4412086110","doi":"10.1016/j.jconhyd.2025.104672","title":"Harnessing deep learning for fusion-based heavy metal contamination index prediction in groundwater","year":2025,"lang":"en","type":"article","venue":"Journal of Contaminant Hydrology","topic":"Geophysical Methods and Applications","field":"Engineering","cited_by":5,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université du Québec en Abitibi-Témiscamingue","funders":"","keywords":"Groundwater; Environmental science; Contamination; Index (typography); Contaminated groundwater; Geology; Environmental remediation; Computer science; Geotechnical engineering; Ecology","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":[0.0005184889,0.0001057802,0.0003102604,0.0002913209,0.00007139553,0.00001928757,0.00009406987,0.0001171434,0.00001085159],"category_scores_gemma":[0.0001068793,0.00009317426,0.00009964903,0.0001878246,0.00003201851,0.0001255904,0.00001069588,0.0003023002,0.000001551156],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009060944,"about_ca_system_score_gemma":0.00003103338,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001060998,"about_ca_topic_score_gemma":0.00003364645,"domain_scores_codex":[0.9990678,0.00009434378,0.0004700059,0.00009851818,0.00008299251,0.0001863208],"domain_scores_gemma":[0.9992813,0.0003716127,0.0001310694,0.00006956823,0.0001124427,0.00003400384],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006094686,0.0005139798,0.04389654,0.0003211506,0.0002265264,0.00002971487,0.0006740184,0.2737806,0.1953456,0.004687418,0.0001098153,0.4798052],"study_design_scores_gemma":[0.002112505,0.000396792,0.1716419,0.0001018402,0.00008682862,0.00001547418,0.0001020192,0.8085656,0.004712865,0.002621475,0.009527521,0.0001151993],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5377535,0.0001317611,0.4613166,0.0002276132,0.0002884529,0.000116685,5.819162e-7,0.00001586251,0.0001489917],"genre_scores_gemma":[0.9976746,0.00001361774,0.001984108,0.00006724112,0.00009426721,0.0000405894,0.000005056419,0.00001183865,0.0001086371],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.534785,"threshold_uncertainty_score":0.3799536,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007874952239483606,"score_gpt":0.255346592497522,"score_spread":0.2474716402580383,"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."}}