{"id":"W4251683700","doi":"10.23970/ahrqepcmethodscreatingefficiencies","title":"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":"report","venue":"","topic":"Meta-analysis and systematic reviews","field":"Decision Sciences","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"University of Alberta; Agency for Healthcare Research and Quality; U.S. Department of Health and Human Services","keywords":"Data extraction; Computer science; Relevance (law); Artificial intelligence; Data mining; Interquartile range; Machine learning; Upload; Randomized controlled trial; Medicine; MEDLINE; Surgery","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.6788412,0.003171496,0.0056937,0.01493147,0.002517019,0.01051533,0.006394752,0.003126383,0.006265154],"category_scores_gemma":[0.8604055,0.003981355,0.008686495,0.01714115,0.004414682,0.01556205,0.01223546,0.003583699,0.002976836],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.01014278,"about_ca_system_score_gemma":0.03039494,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001823619,"about_ca_topic_score_gemma":0.002978312,"domain_scores_codex":[0.2449861,0.5733074,0.126274,0.01637005,0.03703253,0.002029942],"domain_scores_gemma":[0.02986408,0.8634214,0.0351694,0.04153787,0.02869913,0.001308159],"domain_codex":"methods","domain_gemma":"methods","domain_candidate":"methods","domain_consensus":"methods","study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.02175696,0.001784565,0.02982016,0.08126778,0.008815818,0.0005487623,0.01089755,0.008609848,0.003664011,0.006798658,0.02758174,0.7984542],"study_design_scores_gemma":[0.07527788,0.03996826,0.130752,0.09648865,0.04384468,0.006175474,0.009928089,0.2262764,0.05594074,0.06454107,0.2458106,0.004996214],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1607035,0.02415591,0.6238725,0.01418794,0.001569783,0.1251712,0.01529653,0.02638469,0.00865801],"genre_scores_gemma":[0.1078993,0.001672039,0.8267781,0.001877952,0.0002549805,0.05796498,0.001827884,0.001308512,0.000416273],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.3211588,"threshold_uncertainty_score":0.396046,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.8885775344023352,"score_gpt":0.6265196369154715,"score_spread":0.2620578974868637,"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."}}