{"id":"W2915915402","doi":"10.14778/3297753.3297758","title":"Cleaning crowdsourced labels using oracles for statistical classification","year":2018,"lang":"en","type":"article","venue":"Proceedings of the VLDB Endowment","topic":"Mobile Crowdsensing and Crowdsourcing","field":"Computer Science","cited_by":36,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Computer science; Oracle; Crowdsourcing; Margin (machine learning); Ground truth; Artificial intelligence; Test data; Data mining; Machine learning; Estimator; Mathematics; Statistics","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":[0.0005551054,0.0001461463,0.0001811678,0.00007237247,0.0004020761,0.0001929467,0.0006990641,0.00005486582,0.000004951262],"category_scores_gemma":[0.000241166,0.0001104984,0.0000768411,0.0002956523,0.0001963291,0.000239247,0.0002698322,0.00009176676,0.000003842712],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000893534,"about_ca_system_score_gemma":0.00004377271,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002536165,"about_ca_topic_score_gemma":0.000001513359,"domain_scores_codex":[0.9986499,0.00001114497,0.0003319565,0.0003672663,0.0003137995,0.0003259314],"domain_scores_gemma":[0.9988934,0.0001041047,0.0002978519,0.0002276108,0.0004086574,0.00006838983],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00003740715,0.00009045936,0.001107606,0.0001149866,0.00003950071,1.823161e-7,0.001961525,0.00003871748,0.7455593,0.2219019,0.00176789,0.0273806],"study_design_scores_gemma":[0.0009097046,0.0002761549,0.003031655,0.0002923075,0.00006621629,0.00003122995,0.0007326252,0.238857,0.7351741,0.01568951,0.004593271,0.0003462305],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.652621,0.00003271144,0.3434343,0.0007782058,0.0004587849,0.0006373984,0.000007686798,0.0001542774,0.001875631],"genre_scores_gemma":[0.875141,0.00000150289,0.1244489,0.0001262303,0.0001634651,0.00001934741,5.074137e-7,0.00001571943,0.00008337833],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2388183,"threshold_uncertainty_score":0.4505996,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04114410296757267,"score_gpt":0.2851606879563389,"score_spread":0.2440165849887662,"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."}}