{"id":"W2604568423","doi":"10.1002/jrsm.1287","title":"Machine learning for identifying Randomized Controlled Trials: An evaluation and practitioner's guide","year":2018,"lang":"en","type":"article","venue":"Research Synthesis Methods","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":449,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"U.S. National Library of Medicine; Agency for Healthcare Research and Quality; National Institutes of Health; National Cancer Institute; Medical Research Council; McMaster University","keywords":"Computer science; Machine learning; Artificial intelligence; Randomized controlled trial; Workflow; Convolutional neural network; Support vector machine; Systematic review; Sensitivity (control systems); Data mining; MEDLINE; Medicine; Database","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":[0.2100705,0.004259774,0.006577351,0.01667193,0.001003564,0.006319019,0.006901622,0.007426151,0.01893827],"category_scores_gemma":[0.5587779,0.003594294,0.005971762,0.01203125,0.003034109,0.005988806,0.003604925,0.01240793,0.01168242],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005679495,"about_ca_system_score_gemma":0.01549808,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006348302,"about_ca_topic_score_gemma":0.00615452,"domain_scores_codex":[0.7991815,0.146843,0.02860472,0.004139943,0.02071245,0.0005184914],"domain_scores_gemma":[0.3672143,0.5356547,0.0136523,0.02065982,0.06001225,0.002806647],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005514781,0.0002847545,0.0009902674,0.02369134,0.001528909,0.0001970213,0.000383651,0.01143105,0.0007069937,0.02734735,0.339139,0.5937482],"study_design_scores_gemma":[0.004148997,0.001284589,0.00335906,0.08204656,0.002672042,0.001440626,0.0003656223,0.1474887,0.003240681,0.2086571,0.5444868,0.0008093728],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0008177368,0.0957329,0.8145073,0.04422575,0.004963846,0.01222513,0.007222296,0.007822282,0.01248283],"genre_scores_gemma":[0.005614884,0.02613099,0.9457957,0.005286744,0.001624823,0.01072935,0.001462274,0.0008843221,0.002471032],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7899294,"threshold_uncertainty_score":0.9741237,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4261457634092655,"score_gpt":0.6226321823758852,"score_spread":0.1964864189666198,"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."}}