{"id":"W2583699183","doi":"","title":"Data Ingestion for the Connected World.","year":2017,"lang":"en","type":"article","venue":"Conference on Innovative Data Systems Research","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":68,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Ingestion; Medicine","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[{"model":"gemma","categories":[],"domain":null,"study_design":"bench_or_experimental","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":[],"domain":null,"study_design":"not_applicable","genre":"software","about_ca_system":false,"about_ca_topic":false,"confidence":"high","status":"direct model label, unvalidated"}],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.008336633,0.0009813579,0.001015283,0.003465313,0.002526005,0.009847531,0.002847907,0.002645242,0.02167739],"category_scores_gemma":[0.02964368,0.0008862456,0.001240175,0.005397003,0.001902352,0.01842966,0.008626588,0.002972838,0.01006711],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001470906,"about_ca_system_score_gemma":0.003430326,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003883541,"about_ca_topic_score_gemma":0.003923412,"domain_scores_codex":[0.9952683,0.001658594,0.0004135694,0.0006789845,0.001659924,0.0003205624],"domain_scores_gemma":[0.9777657,0.006688742,0.000601296,0.01075168,0.003140142,0.001052456],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001537388,0.0003571335,0.01259934,0.0008803311,0.0003752091,0.000926865,0.001316612,0.008563974,0.005718277,0.1195946,0.3800803,0.4680501],"study_design_scores_gemma":[0.0002322377,0.0001991492,0.00607223,0.001023006,0.0002622201,0.001010575,0.002236554,0.1126787,0.02027926,0.270132,0.5857106,0.0001634312],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.05185302,0.0146488,0.6837714,0.0453192,0.0055824,0.001124915,0.02441284,0.09136271,0.08192468],"genre_scores_gemma":[0.4607695,0.01063785,0.4175049,0.004765498,0.002063966,0.0005849376,0.05253784,0.01098053,0.04015487],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02167739,"threshold_uncertainty_score":0.07251805,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.9484017230905639,"score_gpt":0.6713623530691826,"score_spread":0.2770393700213813,"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."}}