{"id":"W4408135179","doi":"10.3389/fdgth.2025.1495040","title":"A simplified retriever to improve accuracy of phenotype normalizations by large language models","year":2025,"lang":"en","type":"article","venue":"Frontiers in Digital Health","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Normalization (sociology); Computer science; Natural language processing; Artificial intelligence; Labrador Retriever; Term (time); Language model; Machine learning; Data mining","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.003418986,0.001956519,0.00143141,0.003099182,0.0005766321,0.002011467,0.00198965,0.001074364,0.009882067],"category_scores_gemma":[0.0177621,0.0005967089,0.001974424,0.002129986,0.0005731731,0.004527699,0.002996106,0.001601343,0.008724117],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000733321,"about_ca_system_score_gemma":0.001700441,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004776557,"about_ca_topic_score_gemma":0.006755498,"domain_scores_codex":[0.9972458,0.0007555132,0.0003567762,0.0006733045,0.0008631602,0.0001054815],"domain_scores_gemma":[0.9933329,0.00340469,0.0004094535,0.001719503,0.00101171,0.0001217109],"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":[0.0006816731,0.0003437647,0.005136684,0.00116634,0.0004141805,0.0006349378,0.0006533476,0.0171664,0.07908715,0.007319309,0.04414374,0.8432525],"study_design_scores_gemma":[0.0003656639,0.0005524046,0.005748602,0.0001661916,0.0005937329,0.002603381,0.000725298,0.7692099,0.1149461,0.02986949,0.07493828,0.0002809003],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04960695,0.001116051,0.8674667,0.0008991166,0.0002437196,0.0004468795,0.006254388,0.06991034,0.004055845],"genre_scores_gemma":[0.1916819,0.0007272126,0.7774184,0.0007228137,0.0002317202,0.0004918444,0.01845482,0.003926233,0.006345036],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009882067,"threshold_uncertainty_score":0.03305882,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007173059222425137,"score_gpt":0.2910441178876113,"score_spread":0.2838710586651862,"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."}}