{"id":"W4408080426","doi":"10.1016/j.geoderma.2025.117223","title":"The problematic case of data leakage: A case for leave-profile-out cross-validation in 3-dimensional digital soil mapping","year":2025,"lang":"en","type":"article","venue":"Geoderma","topic":"Soil Geostatistics and Mapping","field":"Environmental Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ministry of Agriculture, Food and Rural Affairs; Dalhousie University","funders":"Social Sciences and Humanities Research Council of Canada; Agriculture and Agri-Food Canada; Natural Sciences and Engineering Research Council of Canada","keywords":"Leakage (economics); Environmental science; Soil science; Geology; Remote sensing","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.1838913,0.002037616,0.002170447,0.001812955,0.002658958,0.00345651,0.003764461,0.00351108,0.001170775],"category_scores_gemma":[0.2908054,0.0008519085,0.002079257,0.002195397,0.004915693,0.003828143,0.003758717,0.005432493,0.0006511568],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001942113,"about_ca_system_score_gemma":0.003476461,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006881376,"about_ca_topic_score_gemma":0.008272381,"domain_scores_codex":[0.8978451,0.07214262,0.00716019,0.009882353,0.01191828,0.001051525],"domain_scores_gemma":[0.6357345,0.2795384,0.01585901,0.04325994,0.02380705,0.001801137],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003132925,0.00113695,0.1526307,0.00294745,0.004067741,0.002409337,0.007015428,0.2557406,0.01346934,0.04223672,0.02687711,0.4883357],"study_design_scores_gemma":[0.000474236,0.002308142,0.05078471,0.002801505,0.0007290829,0.002206051,0.002560719,0.7189103,0.026126,0.1523958,0.04023962,0.0004638365],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08131497,0.004301517,0.9022517,0.004793892,0.0008140388,0.0007112076,0.0006077387,0.00217124,0.003033803],"genre_scores_gemma":[0.6507128,0.0006173672,0.3371298,0.00566491,0.000368469,0.001251872,0.001459691,0.001060608,0.001734505],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8161087,"threshold_uncertainty_score":0.9725217,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03913837256751391,"score_gpt":0.3139930845743008,"score_spread":0.2748547120067868,"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."}}