{"id":"W2154726829","doi":"10.15398/jlm.v2i1.78","title":"Evaluation of automatic updates of Roget’s Thesaurus","year":2014,"lang":"en","type":"article","venue":"Journal of Language Modelling","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Thesaurus; Computer science; Information retrieval; Natural language processing; WordNet; Vocabulary; Artificial intelligence; Word (group theory); Linguistics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004144822,0.00009148331,0.0002773034,0.0002302927,0.00002310152,0.00003259806,0.0007039059,0.00005479737,0.00001269546],"category_scores_gemma":[0.0003057778,0.00006791489,0.0001040762,0.0002186445,0.00002695799,0.0004454048,0.00006780821,0.0001544139,7.454004e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004156646,"about_ca_system_score_gemma":0.0001123734,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001995657,"about_ca_topic_score_gemma":0.000001169335,"domain_scores_codex":[0.9980474,0.0002081139,0.0005675658,0.00009923137,0.0009618856,0.0001157818],"domain_scores_gemma":[0.997761,0.0001354671,0.0009190122,0.0002915827,0.0008537243,0.00003920483],"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":[0.00001925625,0.0002067925,0.0001321039,0.0003431168,0.0001533864,0.00001905785,0.01854353,0.1246304,0.07830291,0.01673875,0.0000855675,0.7608252],"study_design_scores_gemma":[0.0002010578,0.00007628333,0.000005660032,0.0002428912,0.00005933995,0.00003273358,0.00008395118,0.8357712,0.1318402,0.03162175,0.000004678109,0.00006020657],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2639205,0.005059557,0.7306725,0.0000615986,0.00006760855,0.0000506405,4.275842e-7,0.00002974761,0.0001374359],"genre_scores_gemma":[0.602397,0.000009416543,0.3975336,0.00001653571,0.00003466072,5.169632e-7,2.239118e-7,0.000004828696,0.000003242762],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.760765,"threshold_uncertainty_score":0.2769489,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02970266816764226,"score_gpt":0.3008975775526793,"score_spread":0.271194909385037,"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."}}