{"id":"W3014082802","doi":"10.2196/12799","title":"Identification of the Best Semantic Expansion to Query PubMed Through Automatic Performance Assessment of Four Search Strategies on All Medical Subject Heading Descriptors: Comparative Study","year":2020,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Unified Medical Language System; Computer science; Information retrieval; Synonym (taxonomy); Precision and recall; Search engine indexing; Subject (documents); Construct (python library); Term (time); Query expansion; Semantics (computer science); Measure (data warehouse); Index (typography); Data mining; World Wide Web","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":[0.001058643,0.0001806518,0.0003974446,0.00004898145,0.00008122927,0.00003118574,0.0006517472,0.0002364401,0.00002328269],"category_scores_gemma":[0.0004756333,0.0001172272,0.00008762899,0.0003027934,0.0003452278,0.00002414622,0.0003076504,0.0003604492,0.000007671741],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003207813,"about_ca_system_score_gemma":0.0004939315,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001803946,"about_ca_topic_score_gemma":0.00001628638,"domain_scores_codex":[0.9967196,0.0002081175,0.001001074,0.0001785557,0.001595629,0.0002970518],"domain_scores_gemma":[0.998866,0.00009868423,0.0002798369,0.0003594884,0.0001330283,0.0002629712],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001284265,0.0113329,0.180131,0.01392781,0.003091442,0.00006002417,0.366358,0.003601707,0.07375389,0.001445135,0.03199344,0.3130204],"study_design_scores_gemma":[0.005777786,0.01314464,0.2968956,0.002469269,0.0002521596,0.00006430739,0.2452923,0.3210574,0.1106204,0.00006677756,0.003150089,0.001209312],"study_design_candidate":"qualitative","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9932443,0.00002640912,0.004296149,0.00115551,0.000162954,0.0007973066,0.000007612362,0.00002363528,0.0002860989],"genre_scores_gemma":[0.9983848,0.0000391572,0.0006366038,0.0006879078,0.00009825967,0.0001127056,0.00001666368,0.000009196613,0.00001474217],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3174557,"threshold_uncertainty_score":0.4780384,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09308255225453313,"score_gpt":0.3844331716753633,"score_spread":0.2913506194208302,"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."}}