{"id":"W6927714599","doi":"10.3389/fenvs.2025.1543852.s001","title":"Supplementary file 1_Regression-based machine learning models for nitrate and chloride prediction in surface water in a small agricultural sand plain sub-watershed in southwestern Ontario, Canada.docx","year":2025,"lang":"en","type":"dataset","venue":"Figshare","topic":"Microbial Natural Products and Biosynthesis","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Surface water; Water quality; Regression; Hydrology (agriculture); Sampling (signal processing); Regression analysis; Support vector machine; Nitrate; Agriculture","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0009180799,0.000796443,0.0007801006,0.001498257,0.001089734,0.001513322,0.00253411,0.0008286471,0.4324251],"category_scores_gemma":[0.00826329,0.0004400151,0.0007192715,0.003950053,0.0003069051,0.0008640531,0.000634383,0.0006746793,0.0513581],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.006533951,"about_ca_system_score_gemma":0.01293193,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.7968442,"about_ca_topic_score_gemma":0.8979537,"domain_scores_codex":[0.9996907,0.00002316935,0.00003535135,0.00008567195,0.000107309,0.00005791649],"domain_scores_gemma":[0.9954833,0.001456348,0.0002063478,0.0001641089,0.00251034,0.0001795905],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0000978857,0.00004415257,0.01388317,0.001397407,0.00004088542,0.0001398781,0.0001664931,0.004574071,0.0002498162,0.0005372519,0.9583227,0.02054642],"study_design_scores_gemma":[0.0009236114,0.0001012867,0.1117637,0.00137862,0.0001378696,0.0002636686,0.001335247,0.04563191,0.001909573,0.002969662,0.8333908,0.000194075],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.001442261,0.00003189846,0.0007231043,0.0001518725,0.00003477562,0.00006962546,0.9955848,0.0004799354,0.001481679],"genre_scores_gemma":[0.02448729,0.0002576371,0.01103003,0.0002279801,0.0000466168,0.0006390818,0.9457336,0.0005897793,0.01698797],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.4324251,"threshold_uncertainty_score":0.8095767,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02007649384832817,"score_gpt":0.2112608072518074,"score_spread":0.1911843134034792,"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."}}