{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001627182,0.001104533,0.0009338831,0.0009791049,0.0006566966,0.001219603,0.002262704,0.001590809,0.004655684],"category_scores_gemma":[0.003862886,0.0004301495,0.001090143,0.0007080835,0.0003307885,0.002026315,0.001299471,0.00134887,0.002819682],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001191854,"about_ca_system_score_gemma":0.001742857,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02113808,"about_ca_topic_score_gemma":0.02512093,"domain_scores_codex":[0.9990739,0.0001861708,0.00008980492,0.0003795953,0.0002013,0.00006922057],"domain_scores_gemma":[0.9982134,0.0009943444,0.00007119486,0.0002007183,0.0004305737,0.00008980993],"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":[0.001596895,0.001422875,0.01446451,0.0004615406,0.000522771,0.001565346,0.0004496012,0.1730964,0.02187483,0.004212488,0.03884722,0.7414854],"study_design_scores_gemma":[0.00005437808,0.0001017904,0.001071796,0.00001755297,0.0000655751,0.0001329347,0.00003245763,0.9886818,0.005206849,0.001810112,0.002802653,0.00002208209],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1262374,0.001037641,0.8098772,0.001620774,0.0003391642,0.001016598,0.006797953,0.04897182,0.004101467],"genre_scores_gemma":[0.3277203,0.000424059,0.6462697,0.001233191,0.0001009259,0.001013139,0.01239613,0.0006084826,0.01023407],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02113808,"threshold_uncertainty_score":0.0420301,"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."}}