{"id":"W3017538732","doi":"10.2196/17638","title":"Document-Level Biomedical Relation Extraction Using Graph Convolutional Network and Multihead Attention: Algorithm Development and Validation","year":2020,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Natural Science Foundation of China","keywords":"Computer science; Relationship extraction; Artificial intelligence; Word embedding; Dependency graph; Graph; Natural language processing; Information extraction; Precision and recall; Convolutional neural network; Dependency (UML); Context (archaeology); Embedding; Theoretical computer science","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":[0.002205786,0.002016709,0.001099371,0.002217816,0.0007030111,0.001014268,0.002574571,0.002178346,0.002470225],"category_scores_gemma":[0.004369531,0.0006520796,0.001216091,0.001600044,0.0006070319,0.001799578,0.001166313,0.00194704,0.00095263],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00288258,"about_ca_system_score_gemma":0.003270362,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.04781404,"about_ca_topic_score_gemma":0.0466756,"domain_scores_codex":[0.9991954,0.0001520795,0.00006462028,0.000270239,0.0001870215,0.0001306474],"domain_scores_gemma":[0.9978967,0.0009627688,0.00015045,0.0002648919,0.000653712,0.00007143693],"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":[0.0004925505,0.0006210197,0.006991054,0.0004140717,0.0003954532,0.0003548378,0.00008790041,0.2689268,0.01401712,0.001728853,0.008582491,0.6973878],"study_design_scores_gemma":[0.00003081662,0.00005399894,0.0007578919,0.00001177489,0.00003778898,0.00004112656,0.00001381942,0.99362,0.004319339,0.0006296451,0.000475122,0.000008761253],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.3192506,0.006170419,0.6373422,0.001156342,0.0003553856,0.001283445,0.002261117,0.02623957,0.005940896],"genre_scores_gemma":[0.5716228,0.001393439,0.4121308,0.0005491211,0.00007260583,0.0007711301,0.007612621,0.0003736399,0.005473853],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.04781404,"threshold_uncertainty_score":0.09507149,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.040842741485564,"score_gpt":0.314759726831804,"score_spread":0.27391698534624,"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."}}