{"id":"W2994804941","doi":"10.1016/j.jbi.2019.103354","title":"Task definition, annotated dataset, and supervised natural language processing models for symptom extraction from unstructured clinical notes","year":2019,"lang":"en","type":"article","venue":"Journal of Biomedical Informatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":35,"is_retracted":false,"has_abstract":false,"ca_institutions":"Health Sciences Centre; University of Toronto; Sunnybrook Health Science Centre","funders":"","keywords":"Coreference; Computer science; Natural language processing; Artificial intelligence; Named-entity recognition; Recall; Normalization (sociology); Task (project management); Automatic summarization; Information extraction; Annotation; Relationship extraction; Precision and recall; F1 score; Language model; Machine learning; Information retrieval; Resolution (logic)","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009219837,0.0001318778,0.00033336,0.0001717584,0.00009038362,0.0001900279,0.0004622622,0.0001674115,0.00001010854],"category_scores_gemma":[0.0005689301,0.00009700515,0.00007154935,0.0002178525,0.00008385938,0.001626467,0.0001081643,0.0006357186,0.000004886905],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003530232,"about_ca_system_score_gemma":0.0002269867,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002375244,"about_ca_topic_score_gemma":0.000002197863,"domain_scores_codex":[0.9978251,0.00008569146,0.001258894,0.0001135803,0.0005019436,0.000214757],"domain_scores_gemma":[0.9976167,0.0008377658,0.0008241971,0.0002273237,0.0002672806,0.000226699],"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":[0.0002840891,0.0001925735,0.005151092,0.00121154,0.000170371,0.00003601676,0.01531526,0.0007249848,0.000890062,0.000981241,0.005665812,0.969377],"study_design_scores_gemma":[0.001405688,0.0002805543,0.002572946,0.0001920205,0.00002262497,0.0001613031,0.0004071631,0.9911417,0.00003002668,0.001812958,0.001840814,0.0001322009],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4668325,0.0006330075,0.5293593,0.001601301,0.001059473,0.0002248411,0.0002307576,0.00004172656,0.00001711311],"genre_scores_gemma":[0.739187,0.00004328315,0.2592177,0.0009383248,0.0002007907,0.000001244158,0.0004028846,0.000006957453,0.000001840785],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9904167,"threshold_uncertainty_score":0.3955755,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03168750268367153,"score_gpt":0.350800854451614,"score_spread":0.3191133517679424,"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."}}