{"id":"W3043267198","doi":"10.48550/arxiv.2007.09141","title":"Diversifying Anonymized Data with Diversity Constraints","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University; Western University","funders":"","keywords":"Diversity (politics); Computer science; Variety (cybernetics); Cluster analysis; Data science; Set (abstract data type); Data set; Data mining; Information retrieval; Artificial intelligence; Political science; Law","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":[0.0127371,0.0008945628,0.001916568,0.002425435,0.002792106,0.006138377,0.00341108,0.002700765,0.001477533],"category_scores_gemma":[0.05051611,0.001006458,0.002408027,0.005886817,0.003737515,0.011059,0.007468218,0.005939881,0.0004439358],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002712744,"about_ca_system_score_gemma":0.0040624,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001626389,"about_ca_topic_score_gemma":0.002000608,"domain_scores_codex":[0.9785399,0.009025634,0.001742835,0.004042619,0.005492886,0.001156114],"domain_scores_gemma":[0.9419498,0.03495234,0.004831146,0.01444429,0.002864876,0.0009575869],"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":[0.0007631532,0.0004095313,0.01005182,0.0004391426,0.0003161795,0.001141027,0.002231493,0.4487745,0.007910502,0.3674092,0.007945952,0.1526077],"study_design_scores_gemma":[0.00007162288,0.00008472722,0.0005973981,0.00008212951,0.00005986885,0.0006618418,0.0005767714,0.5072425,0.008457839,0.4750365,0.007078546,0.00005030579],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0765429,0.000541357,0.9133521,0.003602815,0.00007206009,0.000316438,0.001318012,0.0005979432,0.003656423],"genre_scores_gemma":[0.5168587,0.0005236458,0.4757125,0.00108162,0.0001783313,0.000365198,0.002763494,0.0002180686,0.002298519],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0127371,"threshold_uncertainty_score":0.067361,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2088540806123089,"score_gpt":0.2248576007321856,"score_spread":0.01600352011987669,"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."}}