{"id":"W6976885201","doi":"10.6084/m9.figshare.5861067.v1","title":"High-definition phonotactic typology in Sahul: Choosing the data-rich approach","year":2018,"lang":"en","type":"article","venue":"Figshare","topic":"Nuclear Engineering Thermal-Hydraulics","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Phonotactics; Typology; Linguistic typology; Variation (astronomy); Phonology; Lexical diversity; Rule-based machine translation; Diversity (politics)","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":["insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.0000733378,0.0001383984,0.0001248286,0.00007147003,0.00004803974,0.00004369701,0.0004813244,0.0001053728,0.009932443],"category_scores_gemma":[0.000229631,0.0001221407,0.00001633457,0.0002578184,0.00001135817,0.0001873438,0.0001544789,0.0002854152,0.003181239],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006756993,"about_ca_system_score_gemma":0.00001178604,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001055516,"about_ca_topic_score_gemma":0.00002216646,"domain_scores_codex":[0.9992721,0.00002632446,0.0001468528,0.0001930496,0.00009159852,0.0002700418],"domain_scores_gemma":[0.9991927,0.00008543893,0.0000231431,0.0006410392,0.00002292192,0.00003474307],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003080766,0.0001260985,0.00005507178,0.0007170851,0.0001610823,0.00004120975,0.00181075,0.07652052,0.002987809,0.001007842,0.9020574,0.01448434],"study_design_scores_gemma":[0.00117048,0.000094704,0.03087954,0.0008443373,0.0000467482,0.0001411611,0.0002265654,0.6441802,0.003688499,0.0005270994,0.3169851,0.001215616],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.3380638,0.004025659,0.001851263,0.001407254,0.006707609,0.003891147,0.2142462,0.01083074,0.4189763],"genre_scores_gemma":[0.9812248,0.00000437594,0.001367949,0.0001030365,0.000411337,0.00003962233,0.01676198,0.00007367082,0.00001327214],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6431609,"threshold_uncertainty_score":0.9975949,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06409533963910465,"score_gpt":0.2346857216734122,"score_spread":0.1705903820343075,"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."}}