{"id":"W4318187137","doi":"10.1109/bigdata55660.2022.10020293","title":"Entity Matching with AUC-Based Fairness","year":2022,"lang":"en","type":"article","venue":"2022 IEEE International Conference on Big Data (Big Data)","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University","funders":"","keywords":"Computer science; Matching (statistics); Statistics; Mathematics","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":[0.02678535,0.001245359,0.002721226,0.003180128,0.002043812,0.005003613,0.003926276,0.00319431,0.002434625],"category_scores_gemma":[0.1364511,0.0006941913,0.001489522,0.004596928,0.003615809,0.008819884,0.006642721,0.003006437,0.0008215393],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003000777,"about_ca_system_score_gemma":0.003236826,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003730386,"about_ca_topic_score_gemma":0.002042626,"domain_scores_codex":[0.9803161,0.009162597,0.001278789,0.005053448,0.003230415,0.0009585649],"domain_scores_gemma":[0.9307194,0.04092634,0.004897801,0.01620779,0.005929904,0.001318798],"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":[0.0008998867,0.0002436894,0.02193455,0.0002381851,0.0002861011,0.0002996953,0.000722184,0.5717043,0.002300524,0.1328978,0.005417662,0.2630554],"study_design_scores_gemma":[0.00002857802,0.00008540067,0.001347073,0.00002921148,0.00003251544,0.0001679258,0.00006326329,0.8687935,0.003171737,0.1237694,0.002472335,0.00003892709],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03713205,0.0008125136,0.9573455,0.0007539403,0.00009623414,0.0001486396,0.0002644452,0.001055056,0.002391632],"genre_scores_gemma":[0.7072269,0.0003566648,0.2873195,0.000518674,0.0002046467,0.0002284506,0.0007127209,0.0003759531,0.003056445],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02678535,"threshold_uncertainty_score":0.1416562,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.6664563814827509,"score_gpt":0.4557380255016488,"score_spread":0.2107183559811021,"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."}}