{"id":"W4205570006","doi":"10.1145/3492853","title":"Studying Up Machine Learning Data","year":2022,"lang":"en","type":"article","venue":"Proceedings of the ACM on Human-Computer Interaction","topic":"Ethics and Social Impacts of AI","field":"Social Sciences","cited_by":113,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"International Development Research Centre","keywords":"Documentation; Transparency (behavior); Computer science; Data science; Context (archaeology); Argument (complex analysis); Perspective (graphical); Quality (philosophy); Data quality; Power (physics); Work (physics); Artificial intelligence; Machine learning; Knowledge management; Epistemology; Engineering; Business; Computer security; Marketing","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["sts"],"consensus_categories":[],"category_scores_codex":[0.001960057,0.0001055599,0.000153276,0.00009910642,0.002808519,0.0002714099,0.002932293,0.00004727883,0.0001635142],"category_scores_gemma":[0.001349239,0.00009235892,0.00008068467,0.0002554582,0.0001044533,0.0008326239,0.002993332,0.0009721685,0.000005865549],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001989914,"about_ca_system_score_gemma":0.00004074021,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001609461,"about_ca_topic_score_gemma":0.0001213152,"domain_scores_codex":[0.9984381,0.00008626773,0.0002381127,0.000291342,0.0007248712,0.0002212644],"domain_scores_gemma":[0.9987753,0.0002096528,0.0003837734,0.000347552,0.0002343052,0.00004938867],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0004591891,0.001573234,0.03915497,0.0002248747,0.0006216763,0.00000405118,0.4540564,0.001359486,0.01930927,0.2063175,0.1871001,0.08981928],"study_design_scores_gemma":[0.001915716,0.001864632,0.01093914,0.0004486626,0.0002003585,0.00001600059,0.1003932,0.01243987,0.003392405,0.09226494,0.7749289,0.001196146],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.959745,0.0000263237,0.00001274269,0.01590208,0.002977336,0.0003309807,0.00001055199,0.0001214687,0.02087349],"genre_scores_gemma":[0.9966181,0.00001741608,0.0002706888,0.0006043179,0.0007404934,0.000009054992,0.000007899697,0.00001617365,0.001715848],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5878288,"threshold_uncertainty_score":0.9984897,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2530857386306079,"score_gpt":0.4353445862956434,"score_spread":0.1822588476650355,"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."}}