{"id":"W2530130430","doi":"","title":"Data quality through active constraint discovery and maintenance","year":2012,"lang":"en","type":"dissertation","venue":"","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Leverage (statistics); Constraint (computer-aided design); Data integrity; Data quality; Data mining; Relation (database); Quality (philosophy); Set (abstract data type); Domain (mathematical analysis); Machine learning; Engineering; Mathematics; Database","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":[0.02900548,0.002241037,0.002464637,0.007035648,0.002871559,0.009496144,0.01000253,0.002591074,0.003469865],"category_scores_gemma":[0.1378233,0.001930361,0.003175421,0.006498178,0.003753225,0.0113581,0.00888131,0.005371659,0.001631175],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003473921,"about_ca_system_score_gemma":0.008437449,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01298418,"about_ca_topic_score_gemma":0.01035367,"domain_scores_codex":[0.9678967,0.00946126,0.003029732,0.005650813,0.01291854,0.001042971],"domain_scores_gemma":[0.8179069,0.08899397,0.01514701,0.05022655,0.02622786,0.001497678],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003900967,0.0005166106,0.01327187,0.001028131,0.0005141852,0.0003439531,0.001629396,0.1012118,0.007925583,0.0700411,0.01626461,0.7868626],"study_design_scores_gemma":[0.0001233168,0.0001315901,0.001657572,0.0002858083,0.0002430156,0.0004805742,0.0006567336,0.8399148,0.0259529,0.1021676,0.02826807,0.0001180333],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.008088269,0.0004597892,0.9851713,0.001096497,0.00007261259,0.0002353402,0.0003287297,0.003191484,0.001356008],"genre_scores_gemma":[0.09452736,0.000410864,0.9009813,0.000381431,0.00008745268,0.0002708676,0.001234263,0.0005642779,0.001542137],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02900548,"threshold_uncertainty_score":0.1533975,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4121116768603989,"score_gpt":0.5156160078328202,"score_spread":0.1035043309724213,"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."}}