{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01724999,0.00282653,0.003594328,0.002764346,0.00222247,0.003571476,0.004950933,0.003006696,0.004417261],"category_scores_gemma":[0.07450181,0.001418273,0.002508818,0.002570615,0.004006381,0.005813381,0.00705739,0.006403655,0.002411463],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002727777,"about_ca_system_score_gemma":0.003923355,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007492248,"about_ca_topic_score_gemma":0.0092473,"domain_scores_codex":[0.9835334,0.007978245,0.0009056167,0.003083444,0.003760756,0.0007384848],"domain_scores_gemma":[0.948283,0.02805463,0.003525972,0.01327551,0.005623104,0.001237712],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001326185,0.0004272452,0.01161864,0.0007614944,0.0002989742,0.0003726961,0.001355676,0.4470429,0.008268123,0.06215555,0.02105194,0.4453207],"study_design_scores_gemma":[0.00005555307,0.00008120497,0.0006638672,0.00005297545,0.00002937201,0.00005612837,0.0001166485,0.9328382,0.00418343,0.05791842,0.003959178,0.00004510475],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.007256486,0.0003626739,0.986035,0.0006293484,0.0000918671,0.0001187306,0.0002399606,0.004305154,0.0009609102],"genre_scores_gemma":[0.3221174,0.0003205531,0.6688122,0.001182933,0.0002826274,0.0005138009,0.002183347,0.00166612,0.00292108],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01724999,"threshold_uncertainty_score":0.09122771,"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."}}