{"id":"W2159249756","doi":"10.1109/icdm.2001.989536","title":"Subject classification in the Oxford English Dictionary","year":2002,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Mitacs","keywords":"Computer science; Subject (documents); Artificial intelligence; Probabilistic logic; Natural language processing; Weighting; Term (time); Naive Bayes classifier; Hierarchy; Word (group theory); Information retrieval; Linguistics; Support vector machine","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.004162904,0.001022014,0.001863969,0.03377176,0.001913586,0.006883719,0.001364439,0.001224146,0.08578789],"category_scores_gemma":[0.02004262,0.0006650898,0.0006803835,0.04429017,0.001519662,0.006891315,0.002742603,0.001754491,0.07326682],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002329009,"about_ca_system_score_gemma":0.007236514,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0092221,"about_ca_topic_score_gemma":0.008520897,"domain_scores_codex":[0.9922516,0.001140916,0.003807163,0.0007672612,0.001774246,0.0002588039],"domain_scores_gemma":[0.9820809,0.005358666,0.002485588,0.002435814,0.006519615,0.001119518],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001079112,0.00003648833,0.002237085,0.004661013,0.00003391014,0.0002773862,0.001335547,0.0003777692,0.001524735,0.04447778,0.722361,0.2225694],"study_design_scores_gemma":[0.000008248393,0.0000102724,0.002374793,0.0005753734,0.000007286801,0.0001448788,0.000207682,0.0001546252,0.0001509956,0.003411182,0.9929364,0.00001818665],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"empirical","genre_scores_codex":[0.007349362,0.04735075,0.05386983,0.008450951,0.006814397,0.001795995,0.5845681,0.007127826,0.2826728],"genre_scores_gemma":[0.03734468,0.0437565,0.1297837,0.002479603,0.002589004,0.001996274,0.6557578,0.003749855,0.1225426],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.08578789,"threshold_uncertainty_score":0.2869891,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02662954429509814,"score_gpt":0.2525118843802993,"score_spread":0.2258823400852012,"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."}}