{"id":"W2950021574","doi":"10.1093/bioinformatics/btz504","title":"Towards reliable named entity recognition in the biomedical domain","year":2019,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Topic Modeling","field":"Computer Science","cited_by":88,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"National Human Genome Research Institute; Compute Canada; National Institutes of Health; Nvidia","keywords":"CRFS; Conditional random field; Computer science; Artificial intelligence; Dropout (neural networks); Transfer of learning; Named-entity recognition; Natural language processing; Machine learning; Task (project management); Regularization (linguistics); Deep learning; Sequence labeling; Source code; Generalization; Multi-task learning","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":[0.0129524,0.002213101,0.001525546,0.004336638,0.000815239,0.002995015,0.00309005,0.003548447,0.006058088],"category_scores_gemma":[0.03567348,0.0007922073,0.001460393,0.003850201,0.001262279,0.009502488,0.004716083,0.003740699,0.01196451],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009554657,"about_ca_system_score_gemma":0.002334482,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003111565,"about_ca_topic_score_gemma":0.002687576,"domain_scores_codex":[0.9939559,0.002384263,0.0004464016,0.001665268,0.001294269,0.0002540308],"domain_scores_gemma":[0.9768441,0.01169566,0.001613877,0.005025121,0.004181539,0.0006396629],"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":[0.001195152,0.0004039822,0.00914653,0.002971592,0.0003875056,0.0007629974,0.00062363,0.1161783,0.03430507,0.02638712,0.1359556,0.6716825],"study_design_scores_gemma":[0.00008797746,0.0001867872,0.003590097,0.0003895955,0.0001364679,0.0006065902,0.0003821173,0.8132056,0.06442299,0.04937023,0.06749763,0.0001238581],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03625008,0.006362987,0.876975,0.004993779,0.0007837351,0.0003014441,0.01591927,0.05203746,0.006376174],"genre_scores_gemma":[0.2205704,0.002569509,0.7038203,0.001404495,0.0005146472,0.0003308952,0.06358568,0.001826252,0.005377881],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0129524,"threshold_uncertainty_score":0.06849968,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02171330360008556,"score_gpt":0.2395539452077731,"score_spread":0.2178406416076875,"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."}}