{"id":"W2460936583","doi":"10.1145/2911451.2914709","title":"Impact of Review-Set Selection on Human Assessment for Text Classification","year":2016,"lang":"en","type":"article","venue":"","topic":"Machine Learning and Algorithms","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Selection (genetic algorithm); Relevance (law); Context (archaeology); Computer science; Sampling (signal processing); Set (abstract data type); Information retrieval; Data mining; Statistics; Machine learning; Artificial intelligence; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003292499,0.00006294966,0.0001001187,0.00004934413,0.00005988219,0.00001677431,0.0001954729,0.00002027149,0.00007352638],"category_scores_gemma":[0.00003965796,0.0000335606,0.00008158471,0.0001250236,0.000008543459,0.0001108268,0.00001796506,0.00003931481,0.0000151828],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006359395,"about_ca_system_score_gemma":0.0000431323,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003251668,"about_ca_topic_score_gemma":0.000002003703,"domain_scores_codex":[0.9993984,0.00005128326,0.0001500687,0.0001800417,0.0001208676,0.00009939274],"domain_scores_gemma":[0.9994742,0.00007957736,0.0001113345,0.000222324,0.00008236337,0.00003018954],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.00000489076,0.0001659683,0.0102678,0.0001140688,0.00003301895,1.094329e-7,0.00003217075,0.00003082851,0.03792887,0.1367217,0.03440662,0.7802939],"study_design_scores_gemma":[0.00140514,0.004387027,0.7847052,0.001027415,0.00002379106,0.000007963356,0.000004040873,0.1848509,0.004163008,0.006002717,0.01298803,0.0004347429],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.008223825,0.00002482915,0.9848262,0.001924135,0.00005967052,0.0002191542,0.00000249818,0.00008341835,0.004636272],"genre_scores_gemma":[0.9730423,0.00004892618,0.02458305,0.0001123501,0.00005474396,0.00002986474,0.000004229254,0.000004798947,0.002119736],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9648185,"threshold_uncertainty_score":0.1368562,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05918482990337513,"score_gpt":0.4245141933815084,"score_spread":0.3653293634781332,"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."}}