{"id":"W3041938240","doi":"10.1109/msec.2020.2992821","title":"Seven Ways to Evaluate the Utility of Synthetic Data","year":2020,"lang":"en","type":"article","venue":"IEEE Security & Privacy","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":64,"is_retracted":false,"has_abstract":true,"ca_institutions":"Children's Hospital of Eastern Ontario","funders":"","keywords":"Coronavirus disease 2019 (COVID-19); Data sharing; Computer science; Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2); Internet privacy; Medicine","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":["metaresearch","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.001605903,0.0002467187,0.0003514032,0.00006583094,0.0001536584,0.0001583238,0.100547,0.0001160699,0.00007967396],"category_scores_gemma":[0.0394804,0.0001900245,0.00007901548,0.001004243,0.0002189374,0.0009103351,0.1831725,0.0004917228,0.0002039595],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003110459,"about_ca_system_score_gemma":0.0001678921,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009178791,"about_ca_topic_score_gemma":0.00001308325,"domain_scores_codex":[0.9968779,0.0002006956,0.0004899678,0.001166395,0.0007782943,0.0004867242],"domain_scores_gemma":[0.970875,0.0004333931,0.0001396384,0.0282177,0.0001526219,0.0001816974],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0000362322,0.0001585386,0.0003003477,0.0001183219,0.00008490866,0.0000237293,0.00366724,0.000006403502,0.002460742,0.003153185,0.9457743,0.04421605],"study_design_scores_gemma":[0.0002853395,0.0001207063,0.0006806849,0.00005732572,0.00003386313,0.00001251261,0.00005326583,0.5525025,0.03660152,0.383003,0.02630886,0.0003404835],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1064085,0.0003054993,0.6034579,0.2849042,0.0007038595,0.001010219,0.0003897321,0.001203838,0.001616215],"genre_scores_gemma":[0.9368629,0.00003713858,0.06166856,0.001314627,0.00007086265,0.00001505766,0.00001402615,0.00001528532,0.000001494862],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9194654,"threshold_uncertainty_score":0.9686105,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1222913258340352,"score_gpt":0.3242382431896416,"score_spread":0.2019469173556064,"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."}}