{"id":"W1576661435","doi":"","title":"Data quality by design: a goal-oriented approach","year":2010,"lang":"en","type":"dissertation","venue":"TSpace (University of Toronto)","topic":"Software Engineering Research","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Conceptual schema; Quality assurance; Software engineering; Data quality; Data mining; Database; Engineering; Metric (unit)","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":[0.0007640445,0.0002355944,0.0003825606,0.00006941672,0.0001527634,0.00004704409,0.004289758,0.0003907602,0.0007801523],"category_scores_gemma":[0.0003107787,0.0003083402,0.00008351053,0.0001815525,0.0000638678,0.001084238,0.0006921605,0.0004965866,0.00002371337],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002134383,"about_ca_system_score_gemma":0.0002785769,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.05763754,"about_ca_topic_score_gemma":0.0128979,"domain_scores_codex":[0.9978521,0.0001444037,0.0001292728,0.000812655,0.0007163143,0.0003452177],"domain_scores_gemma":[0.9966955,0.0003332125,0.0002215539,0.00235439,0.0002153183,0.0001800051],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.001302275,0.002671293,0.001664259,0.003759127,0.00175542,0.0001536318,0.1269902,0.0002798717,0.04467962,0.01536258,0.2833621,0.5180196],"study_design_scores_gemma":[0.005895214,0.0009124449,0.6756685,0.0005156422,0.0004015729,0.00002243167,0.03756846,0.1804681,0.002720606,0.0001315922,0.09044589,0.005249545],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01511155,0.0009670834,0.9707912,0.00015086,0.0006522709,0.0004829544,0.0001492863,0.0003936066,0.01130122],"genre_scores_gemma":[0.2163429,0.0003873795,0.7107076,0.00001587674,0.0001117998,0.000002876732,0.004636973,0.00008130955,0.06771331],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.6740043,"threshold_uncertainty_score":0.9999369,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03975536040037669,"score_gpt":0.3050509463482356,"score_spread":0.2652955859478589,"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."}}