{"id":"W6969457671","doi":"10.5683/sp3/zcn177","title":"Data Cleaning Sample","year":2023,"lang":"en","type":"dataset","venue":"Borealis","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Windsor","funders":"","keywords":"Sample (material); Sample size determination; Data quality; Data collection; Sampling (signal processing)","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004503005,0.002324353,0.001528501,0.003788903,0.001394659,0.002594902,0.002513943,0.002556698,0.04954226],"category_scores_gemma":[0.03096919,0.0008716904,0.001581721,0.004725112,0.0008994806,0.001183129,0.002768991,0.002386984,0.07788989],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001227593,"about_ca_system_score_gemma":0.00439886,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01367593,"about_ca_topic_score_gemma":0.02380403,"domain_scores_codex":[0.9953347,0.0008718279,0.0006156398,0.00143891,0.001322245,0.0004166486],"domain_scores_gemma":[0.988322,0.002933971,0.0005289515,0.003615429,0.004068861,0.0005308782],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0002001974,0.00008779066,0.001835833,0.0006144042,0.00004719683,0.00005076101,0.00006148047,0.0002854113,0.0006960389,0.0005912421,0.9860195,0.009510222],"study_design_scores_gemma":[0.0004929797,0.00007963486,0.006452149,0.0003283201,0.00006745714,0.0001168744,0.0001941374,0.0008381426,0.002601278,0.002623411,0.9861477,0.00005793703],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.001563961,0.0002137911,0.002224636,0.0002585686,0.0001757508,0.0004499315,0.988519,0.003888247,0.002706146],"genre_scores_gemma":[0.001674211,0.00006429425,0.004264471,0.0002102763,0.00002718628,0.001069585,0.9904796,0.0005815378,0.001628772],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.04954226,"threshold_uncertainty_score":0.1657353,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1231536386978088,"score_gpt":0.3512123089003547,"score_spread":0.2280586702025459,"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."}}