{"id":"W4416942877","doi":"10.1109/bigdata66926.2025.11402455","title":"Relation-Stratified Sampling for Shapley Values Estimation in Relational Databases","year":2025,"lang":"","type":"article","venue":"","topic":"Advanced Database Systems and Queries","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University","funders":"","keywords":"Tuple; Estimator; Joins; Sampling (signal processing); RSS; Relational database; Variance (accounting); Stratification (seeds); Sample (material)","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0009741543,0.0002961281,0.0003648384,0.0004229079,0.0005049507,0.0001168776,0.0002617184,0.0001137613,0.00007735241],"category_scores_gemma":[0.001101869,0.0003044653,0.00009944469,0.0009400593,0.0001032611,0.003281003,0.000253664,0.0002137171,0.00002391524],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001569145,"about_ca_system_score_gemma":0.000562923,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002486192,"about_ca_topic_score_gemma":0.000412133,"domain_scores_codex":[0.9971302,0.000110424,0.001150704,0.0008765057,0.00032418,0.000408005],"domain_scores_gemma":[0.9971713,0.001466273,0.000280191,0.0007626687,0.0002488791,0.00007062498],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003990981,0.00006583173,0.001771932,0.0001731743,0.0000237091,0.000001074348,0.0003090494,0.06179852,0.0001128856,0.9177627,0.0007924475,0.01714878],"study_design_scores_gemma":[0.000850469,0.00004111482,0.01380708,0.0008048426,0.00002053123,0.000002805469,0.0001439542,0.9365069,0.0003256519,0.02438999,0.02277915,0.0003275356],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001597055,0.00072812,0.9906526,0.001272409,0.001484786,0.001109206,0.0002457335,0.0001091735,0.002800985],"genre_scores_gemma":[0.08313483,0.00004726927,0.911642,0.0002213987,0.00008168167,0.0001554817,0.0007191182,0.00001286939,0.003985387],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.8933727,"threshold_uncertainty_score":0.9999408,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08141308089082669,"score_gpt":0.3521301432663022,"score_spread":0.2707170623754754,"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."}}