{"id":"W3129597052","doi":"10.2196/24678","title":"Extracting Drug Names and Associated Attributes From Discharge Summaries: Text Mining Study","year":2021,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":32,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Engineering and Physical Sciences Research Council; King's College London; Saudi Arabian Cultural Bureau","keywords":"CRFS; Computer science; Conditional random field; Named-entity recognition; Natural language processing; Artificial intelligence; Word embedding; Context (archaeology); Biomedical text mining; Relationship extraction; Task (project management); Information extraction; Deep learning; Word (group theory); F1 score; Machine learning; Embedding; Text mining","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003286394,0.0005172263,0.0005017854,0.004361233,0.0004634079,0.0007889888,0.0008972915,0.0009717983,0.0008068844],"category_scores_gemma":[0.01533605,0.0001886522,0.0008959566,0.003367044,0.0003327657,0.001494938,0.0008713952,0.0007040855,0.0007510977],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006490697,"about_ca_system_score_gemma":0.0009779899,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003548526,"about_ca_topic_score_gemma":0.004736715,"domain_scores_codex":[0.997531,0.0005894658,0.0005870507,0.0006744442,0.0005245948,0.00009335593],"domain_scores_gemma":[0.9787053,0.01461779,0.00275827,0.001375807,0.002175038,0.0003676887],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001161606,0.002722614,0.5823756,0.002111627,0.0003880771,0.002840535,0.002031257,0.01095423,0.008752165,0.001161895,0.01546442,0.370036],"study_design_scores_gemma":[0.00033173,0.00168055,0.581966,0.000645031,0.0008902397,0.008392774,0.006290422,0.3096027,0.04641329,0.003399864,0.04016777,0.0002195343],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.9652759,0.001519128,0.01646944,0.0008440287,0.00007142166,0.0004477924,0.0137155,0.0006275048,0.001029246],"genre_scores_gemma":[0.8913017,0.001503466,0.06934205,0.0002612914,0.000155679,0.0003368086,0.03581202,0.00006565818,0.001221343],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.004361233,"threshold_uncertainty_score":0.0173803,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02615739534699034,"score_gpt":0.2886754367506368,"score_spread":0.2625180414036464,"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."}}