{"id":"W4283155630","doi":"10.1145/3531146.3533231","title":"Data Cards: Purposeful and Transparent Dataset Documentation for Responsible AI","year":2022,"lang":"en","type":"article","venue":"2022 ACM Conference on Fairness, Accountability, and Transparency","topic":"Ethics and Social Impacts of AI","field":"Social Sciences","cited_by":183,"is_retracted":false,"has_abstract":true,"ca_institutions":"Google (Canada)","funders":"","keywords":"Documentation; Computer science; Information retrieval; Data science; Artificial intelligence; Programming language","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":["metaresearch","open_science"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.2397558,0.00188919,0.001978709,0.01063099,0.005772913,0.02836437,0.009599047,0.006562269,0.01461389],"category_scores_gemma":[0.4276203,0.004067739,0.002218747,0.009242631,0.007746827,0.04842428,0.03032119,0.01380999,0.01047808],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.006408177,"about_ca_system_score_gemma":0.023807,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005014122,"about_ca_topic_score_gemma":0.005202419,"domain_scores_codex":[0.8159092,0.1051296,0.03818383,0.00974595,0.02818069,0.002850872],"domain_scores_gemma":[0.3489314,0.1853017,0.02738762,0.3707812,0.05883664,0.008761339],"domain_codex":null,"domain_gemma":"reproducibility","domain_candidate":"reproducibility","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0009795137,0.0006561178,0.008341782,0.001757979,0.0002285707,0.001040894,0.027877,0.009287776,0.006188469,0.3828843,0.1748615,0.385896],"study_design_scores_gemma":[0.0002693433,0.0002462535,0.002717725,0.003420248,0.00008685398,0.0005681603,0.004333758,0.02631231,0.008394038,0.2074731,0.7457438,0.0004344443],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.006988659,0.0004798759,0.9272286,0.0165357,0.0007971213,0.002005041,0.005491059,0.0256137,0.01486032],"genre_scores_gemma":[0.05029619,0.0006136718,0.9142381,0.002785268,0.0003773496,0.002740367,0.01275942,0.008386319,0.007803413],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.990401,"threshold_uncertainty_score":0.9375165,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2238368987664745,"score_gpt":0.4491079445249052,"score_spread":0.2252710457584307,"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."}}