{"id":"W2766046864","doi":"10.48550/arxiv.1710.08874","title":"Synthetic Data for Social Good","year":2017,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":50,"is_retracted":false,"has_abstract":true,"ca_institutions":"Impact","funders":"","keywords":"Data science; Computer science","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":["open_science"],"consensus_categories":[],"category_scores_codex":[0.006135399,0.0008101754,0.0005511291,0.002773661,0.001786234,0.003765662,0.001595662,0.001782234,0.02001359],"category_scores_gemma":[0.04511774,0.0003389428,0.00117517,0.003740646,0.00209233,0.003249372,0.004076341,0.002515963,0.003629414],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002523956,"about_ca_system_score_gemma":0.001517037,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003646876,"about_ca_topic_score_gemma":0.005806547,"domain_scores_codex":[0.9915017,0.005298117,0.0003371427,0.0008449794,0.00177555,0.0002425733],"domain_scores_gemma":[0.966614,0.02091262,0.001450138,0.007578567,0.00255038,0.0008943132],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000528859,0.0002827496,0.01337706,0.001102518,0.0001879141,0.0005423157,0.002099389,0.07947461,0.001256193,0.5272434,0.2952376,0.07866742],"study_design_scores_gemma":[0.0001152124,0.0001388245,0.003607541,0.0003702549,0.0000347337,0.0005191844,0.0015326,0.1308834,0.001471382,0.4016898,0.4595425,0.00009467747],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1035874,0.004867991,0.4734604,0.03610464,0.004123958,0.002518324,0.2171036,0.006683799,0.15155],"genre_scores_gemma":[0.565418,0.001547985,0.2782275,0.004353208,0.0007967616,0.003776375,0.1275082,0.001952492,0.01641941],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9984043,"threshold_uncertainty_score":0.06695217,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.220244900746353,"score_gpt":0.2578539553684698,"score_spread":0.03760905462211681,"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."}}