{"id":"W2917277095","doi":"10.1093/jamia/ocy189","title":"deepBioWSD: effective deep neural word sense disambiguation of biomedical text data","year":2018,"lang":"en","type":"article","venue":"Journal of the American Medical Informatics Association","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":39,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa; Dalhousie University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Artificial intelligence; Natural language processing; Initialization; Pipeline (software); Language model; Machine learning; Vocabulary; Word-sense disambiguation; Deep learning; Word (group theory); Artificial neural network; Lexicon; Unified Medical Language System; Classifier (UML)","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":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.001955387,0.0001127316,0.000324193,0.00006796386,0.00007919854,0.00001868179,0.0006635113,0.0001496544,0.0000189896],"category_scores_gemma":[0.008632712,0.00006958535,0.000119228,0.0003234834,0.0007611554,0.00001752947,0.0003597898,0.0002954898,0.000005049843],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000752115,"about_ca_system_score_gemma":0.0001688291,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002045813,"about_ca_topic_score_gemma":0.00002327277,"domain_scores_codex":[0.9976203,0.0002385052,0.0007826079,0.00008571713,0.001060331,0.0002125697],"domain_scores_gemma":[0.9967952,0.0002399964,0.002097387,0.0003660986,0.00034516,0.0001561958],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.000471283,0.0003628838,0.01759749,0.00006792487,0.0008407478,0.000006340996,0.001612293,0.00001846628,0.008752495,0.0000202549,0.03892107,0.9313288],"study_design_scores_gemma":[0.01308735,0.01862367,0.4550851,0.0007675168,0.002588913,0.001358737,0.01613479,0.2522425,0.04571436,0.001272952,0.1911494,0.00197471],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9768693,0.00008062853,0.01968739,0.002532467,0.0005763904,0.00010421,0.00002351763,0.000006211995,0.0001198283],"genre_scores_gemma":[0.9940438,0.00007897722,0.003896929,0.001269305,0.0006349888,0.000001425464,0.00003507337,0.000008085014,0.00003142288],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9293541,"threshold_uncertainty_score":0.999718,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01137229063170583,"score_gpt":0.304357540342778,"score_spread":0.2929852497110721,"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."}}