{"id":"W2606816378","doi":"10.3366/cor.2017.0108","title":"Subtopic annotation and automatic segmentation for news texts in Brazilian Portuguese","year":2017,"lang":"en","type":"article","venue":"Corpora","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Conselho Nacional de Desenvolvimento Científico e Tecnológico; Fundação de Amparo à Pesquisa do Estado de São Paulo","keywords":"Annotation; Computer science; Segmentation; Natural language processing; Portuguese; Artificial intelligence; Process (computing); Rhetorical question; Computational linguistics; Brazilian Portuguese; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000164167,0.00006936518,0.00008854741,0.00007869677,0.0001158602,0.0002387529,0.0003717287,0.00004170463,0.000001268285],"category_scores_gemma":[0.0001093783,0.00006430315,0.00001402174,0.00005696932,0.00003252452,0.0008746284,0.00008893332,0.00004941125,0.000002012756],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002599554,"about_ca_system_score_gemma":0.00003219913,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006164037,"about_ca_topic_score_gemma":0.00007067191,"domain_scores_codex":[0.9994565,0.00001444769,0.0001357508,0.0002003443,0.0000853545,0.0001076084],"domain_scores_gemma":[0.9993628,0.00002888975,0.000195525,0.0003378226,0.00004317405,0.00003180778],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.000003461167,0.00001486328,0.007658435,0.00005979382,0.000001610323,0.00001192283,0.0004075713,3.701718e-7,0.004801407,0.00761992,0.000301563,0.9791191],"study_design_scores_gemma":[0.00138279,0.0002037055,0.4217339,0.0002418631,0.00001053774,0.00002536721,0.00006470013,0.1047649,0.02490793,0.4459456,0.0002627679,0.0004560015],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4951329,0.001279166,0.5006858,0.00148547,0.0002425156,0.0006429128,0.00000258961,0.0003392579,0.0001893662],"genre_scores_gemma":[0.6603045,0.000009340177,0.3394023,0.0001323934,0.00002012517,0.0000309697,0.000005273665,0.000004115213,0.00009101433],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9786631,"threshold_uncertainty_score":0.2622207,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02239593382287337,"score_gpt":0.309573640430716,"score_spread":0.2871777066078426,"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."}}