{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002932304,0.0001583741,0.0003095227,0.00004846494,0.0002592058,0.0002424537,0.0004914349,0.0002657692,0.00001464103],"category_scores_gemma":[0.001560361,0.0001419396,0.00006210348,0.0001573436,0.0000585937,0.0006273402,0.0001131901,0.0004234874,0.00001021204],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000545977,"about_ca_system_score_gemma":0.0005184486,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002331756,"about_ca_topic_score_gemma":0.000001352065,"domain_scores_codex":[0.9972402,0.0002558561,0.001378814,0.0003064987,0.0005178239,0.0003008239],"domain_scores_gemma":[0.9978043,0.0007446202,0.0004130761,0.0002815347,0.0002167057,0.0005398075],"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":[0.0003908174,0.00008777129,0.0002440096,0.0003391847,0.00001166015,3.644671e-7,0.006366207,0.00002185145,0.000003728352,0.002391667,0.0007134652,0.9894293],"study_design_scores_gemma":[0.006630835,0.001742772,0.0007612625,0.00003286881,0.000006168962,0.000002722402,0.000566158,0.9368862,0.00007050311,0.0004719752,0.05266073,0.0001678076],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.08448192,0.00001647219,0.9099414,0.002328858,0.0003038423,0.002629686,0.000004930512,0.0002005656,0.00009226488],"genre_scores_gemma":[0.2670992,0.00003899448,0.7287438,0.001956845,0.0006304142,0.001210552,0.0002823093,0.00001823682,0.0000196682],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9892614,"threshold_uncertainty_score":0.5788129,"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."}}