{"id":"W3012280586","doi":"10.2196/17832","title":"An Ensemble Learning Strategy for Eligibility Criteria Text Classification for Clinical Trial Recruitment: Algorithm Development and Validation","year":2020,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":26,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Science Foundation of Guangdong Province; National Science Foundation","keywords":"Computer science; Artificial intelligence; Machine learning; Gradient boosting; Boosting (machine learning); Classifier (UML); Ensemble learning; F1 score; Encoder; Ensemble forecasting; Random forest","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004788766,0.001230844,0.001731418,0.001806857,0.0005892271,0.0008447716,0.00170273,0.001670662,0.002080732],"category_scores_gemma":[0.009658785,0.000436717,0.001097384,0.0009824033,0.0003174265,0.001363157,0.001200523,0.002267237,0.0009259568],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001177931,"about_ca_system_score_gemma":0.002599421,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008737311,"about_ca_topic_score_gemma":0.009464665,"domain_scores_codex":[0.9986348,0.0005512841,0.0001366021,0.0002678006,0.0002643045,0.000145329],"domain_scores_gemma":[0.9962745,0.002037948,0.000180806,0.0002666801,0.001050535,0.000189629],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004097268,0.0003659047,0.008453728,0.0001098044,0.0002263569,0.0001542968,0.0001304489,0.3566678,0.003214494,0.002293005,0.007687893,0.6202866],"study_design_scores_gemma":[0.0000128085,0.00003234491,0.0002399072,0.000008502011,0.00001459125,0.00001600209,0.000008217539,0.9979748,0.0005110867,0.0009462318,0.0002317435,0.000003783911],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.05780881,0.001058314,0.9348181,0.0008339406,0.0001296383,0.0004093345,0.000378724,0.003414199,0.001149023],"genre_scores_gemma":[0.5571962,0.0005332684,0.4358743,0.000652343,0.0001710071,0.0009120872,0.001791911,0.0001803575,0.002688454],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008737311,"threshold_uncertainty_score":0.02532578,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3118065264941681,"score_gpt":0.5122167515831201,"score_spread":0.200410225088952,"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."}}