{"id":"W2250595697","doi":"","title":"Building NLP resources for Dzongkha: A Tagset and A Tagged Corpus","year":2010,"lang":"en","type":"article","venue":"Publication Server of the Institute for German Language (Institute for German Language)","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National University of Computer and Emerging Sciences; Universität Stuttgart; International Development Research Centre","keywords":"Treebank; Lexicon; Computer science; Natural language processing; Artificial intelligence; Set (abstract data type); Security token; Part of speech; Word (group theory); Training set; Speech recognition; Parsing; Linguistics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"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":[0.001947971,0.0009558507,0.0009596285,0.005148655,0.001727722,0.001988496,0.001594765,0.0008563445,0.01690955],"category_scores_gemma":[0.006235843,0.001290374,0.0007984241,0.003644357,0.0007986526,0.00480283,0.003243754,0.001497638,0.01535249],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001747555,"about_ca_system_score_gemma":0.003854633,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01005164,"about_ca_topic_score_gemma":0.0138845,"domain_scores_codex":[0.9986765,0.0002615418,0.0002031665,0.0004803013,0.0002717553,0.0001066952],"domain_scores_gemma":[0.9965204,0.001378335,0.0003035862,0.000936709,0.0006802716,0.0001806736],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001461633,0.0006312793,0.02167552,0.004158245,0.0002276887,0.00419857,0.00648895,0.02796549,0.08580084,0.06776998,0.2005564,0.5790655],"study_design_scores_gemma":[0.0003056986,0.000208773,0.028443,0.0007204225,0.0002395691,0.002486318,0.00433507,0.1293796,0.06085809,0.0337121,0.7389076,0.0004036998],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"dataset","genre_scores_codex":[0.07072034,0.0007777208,0.6646945,0.001083105,0.0005249132,0.002955888,0.1868358,0.04249397,0.0299137],"genre_scores_gemma":[0.1677539,0.0004885275,0.532663,0.0002221553,0.00006915346,0.003779034,0.2808805,0.003800957,0.01034275],"genre_candidate":"dataset","genre_consensus":null,"teacher_disagreement_score":0.01690955,"threshold_uncertainty_score":0.05656803,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0109800960401046,"score_gpt":0.2980911649840398,"score_spread":0.2871110689439352,"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."}}