{"id":"W4311829616","doi":"10.1371/journal.pdig.0000152","title":"Large-scale application of named entity recognition to biomedicine and epidemiology","year":2022,"lang":"en","type":"article","venue":"PLOS Digital Health","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":43,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University; Public Health Ontario; University of Toronto","funders":"Institute of Health Services and Policy Research; Canadian Institutes of Health Research; University of Toronto","keywords":"Named-entity recognition; Computer science; Biomedicine; Inference; Data science; Parsing; Artificial intelligence; Declaration; Python (programming language); Machine learning; Bioinformatics","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.008179879,0.001650976,0.001110026,0.005279575,0.00110818,0.001804637,0.00274263,0.001912929,0.005316417],"category_scores_gemma":[0.01962097,0.0006335818,0.002532153,0.003924916,0.0007846918,0.004149155,0.003707758,0.002142839,0.004799379],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001440684,"about_ca_system_score_gemma":0.002999826,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007459083,"about_ca_topic_score_gemma":0.008405617,"domain_scores_codex":[0.9946427,0.001879323,0.0007238191,0.001875174,0.0007044645,0.00017449],"domain_scores_gemma":[0.9803056,0.01230445,0.001263118,0.003383999,0.002118915,0.0006239528],"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":[0.0007614766,0.0006114083,0.03606933,0.004060817,0.001008802,0.00324241,0.0009265596,0.05103695,0.02028863,0.009221031,0.177599,0.6951737],"study_design_scores_gemma":[0.0001927022,0.0002989639,0.04010722,0.0007532692,0.000794047,0.004357925,0.0009089381,0.5910508,0.05682712,0.06711933,0.2372457,0.0003439117],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03322249,0.00532114,0.7722021,0.007677366,0.001221399,0.001254706,0.07328825,0.09938613,0.006426431],"genre_scores_gemma":[0.1928073,0.00276028,0.6608959,0.002697525,0.0006663362,0.0008568779,0.1343468,0.001349149,0.003619908],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008179879,"threshold_uncertainty_score":0.0432598,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04028945877442753,"score_gpt":0.3309893893826754,"score_spread":0.2906999306082479,"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."}}