{"id":"W6977112234","doi":"10.6084/m9.figshare.15179126.v1","title":"Additional file 4 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":[{"model":"gemma","categories":["metaresearch"],"domain":"methods","study_design":"simulation_or_modeling","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":["insufficient_payload"],"domain":null,"study_design":"not_applicable","genre":"dataset","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"}],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.01988274,0.001651929,0.002465181,0.006309789,0.000976256,0.003196426,0.003132104,0.001820209,0.8531155],"category_scores_gemma":[0.2569146,0.001364135,0.003241888,0.007371856,0.0006302864,0.00325164,0.002377324,0.001588707,0.0815943],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002281897,"about_ca_system_score_gemma":0.005652322,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003204202,"about_ca_topic_score_gemma":0.006248299,"domain_scores_codex":[0.9905313,0.003490044,0.003214855,0.001128984,0.001241064,0.0003938852],"domain_scores_gemma":[0.5780132,0.3804759,0.01472814,0.01040572,0.01455444,0.001822567],"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":[0.003786224,0.0003091794,0.004753398,0.0488065,0.0008962291,0.0001972485,0.0003058141,0.001700266,0.0003403396,0.003149937,0.8901071,0.04564786],"study_design_scores_gemma":[0.07039201,0.001857864,0.03395042,0.02667706,0.004304079,0.001215834,0.0007436792,0.013742,0.004558234,0.047872,0.7940035,0.000683416],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"empirical","genre_scores_codex":[0.0006140997,0.000091349,0.003846727,0.0004395105,0.00008966044,0.002307582,0.9890003,0.002130585,0.00148012],"genre_scores_gemma":[0.03434401,0.0008033307,0.1260198,0.00261382,0.0006479518,0.09685492,0.7040945,0.007918144,0.02670358],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9801173,"threshold_uncertainty_score":0.2095129,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.82916718861603,"score_gpt":0.5561958584901977,"score_spread":0.2729713301258323,"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."}}