{"id":"W3133042814","doi":"10.2196/22797","title":"A Hybrid Model for Family History Information Identification and Relation Extraction: Development and Evaluation of an End-to-End Information Extraction System","year":2021,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Cancer Institute","keywords":"Computer science; Information extraction; Relationship extraction; Artificial intelligence; Machine learning; Natural language processing; Context (archaeology); Heuristics; Named-entity recognition; Task (project management); Identification (biology); Information retrieval","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002168684,0.0001370351,0.0001800271,0.0002962683,0.000137228,0.0001570631,0.0001791852,0.0001492912,0.000005777812],"category_scores_gemma":[0.0002845802,0.0001443288,0.00002755746,0.0001444762,0.00003082151,0.009313801,0.0000926822,0.0001555919,0.000009544721],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005695113,"about_ca_system_score_gemma":0.0009206331,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007154715,"about_ca_topic_score_gemma":0.000004601282,"domain_scores_codex":[0.9969205,0.00006795487,0.001324466,0.0001282256,0.001395591,0.0001633146],"domain_scores_gemma":[0.9978246,0.00008377402,0.0006234878,0.0003332999,0.0009374693,0.0001973602],"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":[0.00002141433,0.00003958733,0.00001566509,0.000742802,0.00001876398,3.98522e-7,0.0291524,0.00974524,0.0002202641,0.01755068,0.000244639,0.9422482],"study_design_scores_gemma":[0.0006966354,0.00002966152,0.0017074,0.0001016473,0.00002410853,0.00008086453,0.001976288,0.9900891,0.0007194982,0.0002785728,0.00414983,0.0001463927],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3007086,0.00003961806,0.6978209,0.0001243199,0.0002888077,0.0005503964,0.000004776175,0.00007064414,0.0003919635],"genre_scores_gemma":[0.8447442,0.00001309074,0.1544837,0.000240532,0.00003293427,0.0002415551,0.0002220782,0.000004447889,0.00001747356],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9803439,"threshold_uncertainty_score":0.6752281,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05491060037912229,"score_gpt":0.3097503389832227,"score_spread":0.2548397386041004,"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."}}