{"id":"W4320917181","doi":"10.15514/ispras-2022-34(5)-10","title":"Data Mining Methods to Compare Englishes","year":2022,"lang":"en","type":"article","venue":"Proceedings of the Institute for System Programming of RAS","topic":"Linguistics, Language Diversity, and Identity","field":"Arts and Humanities","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"World Englishes; Covert; Noun; Linguistics; Pidgin; Geography","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001252005,0.0001328247,0.0003382482,0.0001144395,0.0007845795,0.0001552476,0.001533868,0.00001472608,0.00001457857],"category_scores_gemma":[0.003296707,0.000109894,0.0001284973,0.0001122471,0.0001926806,0.0002912067,0.00129955,0.00009476331,7.211661e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007487727,"about_ca_system_score_gemma":0.00004588491,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001045976,"about_ca_topic_score_gemma":0.0003944663,"domain_scores_codex":[0.9987662,0.0000154423,0.0003987507,0.000276442,0.0003340287,0.000209076],"domain_scores_gemma":[0.9976891,0.00005677671,0.0004261501,0.0002627664,0.001514446,0.00005074023],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000387809,0.0006393679,0.0210977,0.0130711,0.000911497,0.00000336322,0.4063911,0.000154542,0.0002831094,0.3931226,0.1445267,0.019411],"study_design_scores_gemma":[0.0003695059,0.0001159231,0.0000342939,0.0002703194,0.0002312087,0.000001801523,0.2250336,0.0002920865,0.0009249021,0.0001474418,0.7723734,0.0002055471],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9356285,0.0006282135,0.001236974,0.0001338098,0.04655573,0.003222867,0.002800463,0.0003347742,0.009458655],"genre_scores_gemma":[0.9575137,3.782301e-7,0.03975668,0.0000289069,0.00214153,0.00006933667,0.00004543366,0.0000179451,0.0004260236],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6278467,"threshold_uncertainty_score":0.6034431,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09825657603574751,"score_gpt":0.3175586179290332,"score_spread":0.2193020418932857,"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."}}