{"id":"W6902050021","doi":"10.6084/m9.figshare.15179117.v1","title":"Additional file 1 of Creating efficiencies in the extraction of data from randomized trials: a prospective evaluation of a machine learning and text mining tool","year":2021,"lang":"en","type":"article","venue":"Figshare","topic":"Meta-analysis and systematic reviews","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Information extraction; Extraction (chemistry); Data extraction; Text mining; Training set","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":["metaresearch","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.02056833,0.00156173,0.002319698,0.00612153,0.0009490376,0.002864162,0.00291682,0.001659702,0.8375823],"category_scores_gemma":[0.2678723,0.001342842,0.002834759,0.007285866,0.0006079488,0.003085373,0.002214975,0.001611092,0.07852017],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002132032,"about_ca_system_score_gemma":0.005862862,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002761792,"about_ca_topic_score_gemma":0.005647529,"domain_scores_codex":[0.9906949,0.003587718,0.003091324,0.001113421,0.001165849,0.0003467702],"domain_scores_gemma":[0.5399051,0.4194718,0.01450251,0.01015379,0.01419136,0.00177553],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"observational","study_design_scores_codex":[0.003591804,0.0003232463,0.004538625,0.04622818,0.0008327577,0.0001718429,0.0003113603,0.001700522,0.0003325224,0.00330223,0.890094,0.04857287],"study_design_scores_gemma":[0.07706275,0.00223844,0.03744947,0.02655047,0.004655598,0.001398365,0.0007663072,0.01528268,0.004769142,0.04705392,0.7820596,0.0007132616],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"empirical","genre_scores_codex":[0.0006499311,0.00008977851,0.004277234,0.0004117365,0.00007954207,0.002530674,0.9887257,0.001887964,0.00134745],"genre_scores_gemma":[0.03457806,0.0008361763,0.1412107,0.002457387,0.0006483562,0.1152175,0.675568,0.00710346,0.02238044],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9794317,"threshold_uncertainty_score":0.2316692,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.8310192474067402,"score_gpt":0.556330932377557,"score_spread":0.2746883150291831,"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."}}