{"id":"W4402191750","doi":"10.1080/07268602.2024.2368780","title":"Building a searchable online corpus of Australian and New Zealand aligned speech","year":2024,"lang":"en","type":"article","venue":"Australian Journal of Linguistics","topic":"Linguistic Variation and Morphology","field":"Social Sciences","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Python (programming language); Upload; World Wide Web; Login; Software; The Internet","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":[0.001138858,0.0001213758,0.0003050955,0.0002287861,0.00009973451,0.0001116107,0.000230133,0.0001454805,0.0001968792],"category_scores_gemma":[0.005436757,0.0001122236,0.00009215749,0.0002990192,0.0002433098,0.00004189498,0.00003033241,0.0002578032,0.000005504774],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006703934,"about_ca_system_score_gemma":0.0007757224,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002214021,"about_ca_topic_score_gemma":0.0002629237,"domain_scores_codex":[0.9983924,0.0001203056,0.0006038779,0.0001481897,0.0004140573,0.0003211483],"domain_scores_gemma":[0.9983141,0.0002923028,0.0002746653,0.0001019145,0.000626382,0.0003906746],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0002178627,0.0003118052,0.0187108,0.0003708727,0.0006327138,0.003591776,0.02812812,0.0001388841,0.007240064,0.6417828,0.2779127,0.02096164],"study_design_scores_gemma":[0.0005046171,0.0002632563,0.001050586,0.0003233401,0.0001617104,0.0001362687,0.001018242,0.00002931281,0.00123254,0.02640294,0.9687207,0.0001564601],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8770832,0.001983119,0.0048094,0.06004107,0.03930466,0.00108096,0.0004625233,0.0002876532,0.01494743],"genre_scores_gemma":[0.9234533,0.0002341459,0.03989742,0.00007530309,0.00478378,2.156675e-7,0.000004250861,0.00001936904,0.03153217],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6908081,"threshold_uncertainty_score":0.6508698,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07384442457558994,"score_gpt":0.3845869942070119,"score_spread":0.310742569631422,"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."}}