{"id":"W2770828134","doi":"10.1145/3110025.3120958","title":"datumPIPE","year":2017,"lang":"en","type":"article","venue":"","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"","keywords":"Computer science; Data quality; Data mining; Quality (philosophy); Data integrity; Set (abstract data type); Data set; Database; Artificial intelligence; Engineering","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":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.002231833,0.00002880551,0.00006007171,0.00003193975,0.0003155788,0.001123599,0.001752694,0.0000111184,0.00337024],"category_scores_gemma":[0.002517935,0.00001759397,0.00002509854,0.00002599258,0.00006204714,0.0006186551,0.0006397177,0.00002257269,0.007440115],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000002429436,"about_ca_system_score_gemma":0.000006022173,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001091139,"about_ca_topic_score_gemma":0.0001747587,"domain_scores_codex":[0.9990695,0.00002639144,0.0001404569,0.0001777278,0.0005056998,0.00008028369],"domain_scores_gemma":[0.9979441,0.0001063672,0.00008484661,0.001789213,0.00003628749,0.00003916008],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000001534116,0.00001011756,0.001361019,3.082292e-7,0.000001995807,0.000002650444,0.00002232319,2.879883e-7,0.00001048968,0.2854333,0.5360342,0.1771217],"study_design_scores_gemma":[0.00005800506,0.000004536611,0.04449322,5.677612e-7,9.008312e-7,2.09742e-7,0.0001296121,0.0000844992,0.00009067847,0.08979462,0.8653125,0.0000306648],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.00676987,0.000003871347,0.01759526,0.009830878,0.0004520457,0.0000409638,0.00000767318,0.00002190704,0.9652776],"genre_scores_gemma":[0.8126388,0.000003174499,0.001428647,0.001288934,0.00004598174,0.000001346208,0.000001703419,0.000001302054,0.1845901],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.8058689,"threshold_uncertainty_score":0.9999133,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.644200694101546,"score_gpt":0.5830364078440127,"score_spread":0.06116428625753334,"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."}}