{"id":"W2947100013","doi":"10.1186/s12859-019-2801-x","title":"Automated assessment of biological database assertions using the scientific literature","year":2019,"lang":"en","type":"article","venue":"BMC Bioinformatics","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Bhabha Atomic Research Centre; Australian Research Council","keywords":"Computer science; Consistency (knowledge bases); Information retrieval; Assertion; Relevance (law); Relation (database); Biological database; Classifier (UML); Set (abstract data type); Data mining; Data science; Database; Bioinformatics; Artificial intelligence; Biology; Programming language","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01567042,0.001236855,0.001170785,0.04269748,0.001414736,0.004516311,0.002349785,0.001620355,0.001904236],"category_scores_gemma":[0.08046678,0.0003845561,0.001145705,0.01148153,0.000707281,0.003418221,0.003459313,0.0008554673,0.001377871],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001332849,"about_ca_system_score_gemma":0.004112413,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006640622,"about_ca_topic_score_gemma":0.008072565,"domain_scores_codex":[0.9840577,0.003658361,0.002640794,0.002389656,0.006901947,0.0003515113],"domain_scores_gemma":[0.8685434,0.07798688,0.01400302,0.008757228,0.0288891,0.00182029],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00130869,0.000553153,0.1656629,0.006118868,0.0008405885,0.002061424,0.003432354,0.009749067,0.04501952,0.005748013,0.02945505,0.7300504],"study_design_scores_gemma":[0.000290128,0.0009260915,0.1903131,0.001908668,0.001616992,0.006113408,0.004781835,0.5694304,0.1236708,0.01912922,0.08140146,0.0004178644],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5050017,0.01122569,0.4023477,0.002792213,0.0003590834,0.001829616,0.02946431,0.03805331,0.00892653],"genre_scores_gemma":[0.5306989,0.001196376,0.4334123,0.0003019674,0.0002361773,0.0003301587,0.0323475,0.0004096082,0.001067085],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9843296,"threshold_uncertainty_score":0.08287406,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0457625408629511,"score_gpt":0.3421514158978473,"score_spread":0.2963888750348961,"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."}}