{"id":"W2435238507","doi":"","title":"Data Profiling","year":2018,"lang":"en","type":"book","venue":"","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Profiling (computer programming); Computer science; Metadata; Inference; Data mining; Data science; Artificial intelligence; World Wide Web; Programming language","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":["open_science","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.008580113,0.000153004,0.0002937454,0.0001931332,0.00008287712,0.0005395742,0.006521922,0.0001331275,0.01118233],"category_scores_gemma":[0.002314889,0.0001005787,0.00005113389,0.0001152152,0.0001351642,0.0005116847,0.005522395,0.0001428715,0.03466784],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002932967,"about_ca_system_score_gemma":0.0002510329,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007103308,"about_ca_topic_score_gemma":0.0001464854,"domain_scores_codex":[0.9962528,0.00009617524,0.0006293035,0.001027717,0.001817128,0.0001769082],"domain_scores_gemma":[0.9931091,0.0006811096,0.0002658698,0.005704189,0.0001674615,0.00007224068],"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":[0.000001826053,0.00000603077,8.611706e-7,0.000006355228,0.00001407912,0.000003277637,0.000006668135,6.246282e-8,1.377507e-7,0.1254571,0.8530892,0.0214144],"study_design_scores_gemma":[0.00003533111,0.00001242735,0.000001742166,0.00001300735,0.0000135417,5.357824e-7,0.0000461459,0.0001496292,0.00000299663,0.2017603,0.7978476,0.0001167832],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"other","genre_scores_codex":[2.876995e-7,0.00005877295,0.01970784,0.0004945909,0.0008942983,0.0002187682,0.001614742,0.00006314506,0.9769475],"genre_scores_gemma":[0.000001142516,0.00001810111,0.008696371,0.001359391,0.0006421232,0.000002176915,0.002622035,0.0000119272,0.9866467],"genre_candidate":"other","genre_consensus":"other","teacher_disagreement_score":0.07630312,"threshold_uncertainty_score":0.9988533,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.6744232901337949,"score_gpt":0.5236971452500062,"score_spread":0.1507261448837887,"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."}}