{"id":"W3129829742","doi":"10.1109/icdmw51313.2020.00082","title":"SynC: A Copula based Framework for Generating Synthetic Data from Aggregated Sources","year":2020,"lang":"en","type":"article","venue":"","topic":"Gaussian Processes and Bayesian Inference","field":"Computer Science","cited_by":31,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; sync; Copula (linguistics); Data mining; Scalability; Synthetic data; Machine learning; Artificial intelligence; 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":[0.006346754,0.001252738,0.0007969433,0.002388587,0.0007302319,0.001629905,0.002605995,0.001217534,0.003066659],"category_scores_gemma":[0.02409848,0.0007751764,0.001624398,0.002366578,0.001156404,0.002034453,0.002882187,0.002494698,0.0009244432],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001100458,"about_ca_system_score_gemma":0.001795803,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004909196,"about_ca_topic_score_gemma":0.005498023,"domain_scores_codex":[0.9976934,0.001074725,0.0001480961,0.0004332788,0.0005560564,0.00009446833],"domain_scores_gemma":[0.988856,0.00643216,0.0008605026,0.001995513,0.001558603,0.0002971193],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001606934,0.0001405517,0.005155921,0.0002506609,0.0002076708,0.0002449623,0.0003064438,0.7997056,0.004030981,0.07041127,0.009757618,0.1096278],"study_design_scores_gemma":[0.00001649828,0.00002759006,0.000331062,0.00001412846,0.000007233238,0.00004611816,0.0000238356,0.976996,0.00126896,0.01841901,0.002831268,0.00001822742],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002350374,0.00004174018,0.9955201,0.00009307024,0.00002234691,0.00009825434,0.0005443867,0.0009891378,0.0003405793],"genre_scores_gemma":[0.08361652,0.0001142002,0.9105515,0.0001524796,0.00005526029,0.0006475183,0.003840909,0.0005448505,0.000476675],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006346754,"threshold_uncertainty_score":0.03356522,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07637689567952773,"score_gpt":0.2882800888457438,"score_spread":0.2119031931662161,"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."}}