{"id":"W1851653583","doi":"10.1007/s10994-015-5535-7","title":"Learning to identify relevant studies for systematic reviews using random forest and external information","year":2015,"lang":"en","type":"article","venue":"Machine Learning","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":98,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"","keywords":"Random forest; Computer science; Heuristics; Machine learning; Classifier (UML); Artificial intelligence; Cluster analysis; Systematic review; Class (philosophy); Recall; USable; Task (project management); Data mining; Information retrieval; Natural language processing; World Wide Web","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.1849895,0.005753219,0.01757909,0.04652081,0.002354458,0.006636673,0.005025201,0.004101497,0.007545852],"category_scores_gemma":[0.4840657,0.002292073,0.02496883,0.01792799,0.002084471,0.007863043,0.005558532,0.004211807,0.001685498],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003399175,"about_ca_system_score_gemma":0.01237493,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003095452,"about_ca_topic_score_gemma":0.00919322,"domain_scores_codex":[0.8165774,0.1079019,0.04521941,0.01441708,0.0149835,0.0009006847],"domain_scores_gemma":[0.34864,0.5865281,0.03030043,0.01885713,0.01405245,0.001621852],"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":[0.004777266,0.0009337264,0.02502521,0.2635788,0.08095103,0.001002849,0.001728549,0.02183109,0.0028123,0.008031781,0.02511355,0.5642139],"study_design_scores_gemma":[0.0111278,0.004142092,0.01878856,0.1249988,0.3188915,0.002045193,0.00128715,0.2671984,0.00754222,0.1971214,0.04605855,0.0007983935],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03310554,0.1649452,0.7132946,0.006882929,0.001751527,0.03918669,0.03231756,0.005042647,0.003473354],"genre_scores_gemma":[0.216537,0.01794181,0.7085589,0.00225678,0.001018028,0.03696463,0.01552054,0.0003580817,0.0008442601],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8150105,"threshold_uncertainty_score":0.9783294,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09096737967275106,"score_gpt":0.3815705795832492,"score_spread":0.2906031999104981,"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."}}