{"id":"W4390026249","doi":"10.1353/dic.2023.a915069","title":"Designing Corpus-Creation Tools for Language Revitalization","year":2023,"lang":"en","type":"article","venue":"Dictionaries","topic":"Discourse Analysis in Language Studies","field":"Arts and Humanities","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Documentation; Computer science; Set (abstract data type); Adaptation (eye); Linguistics; Spoken language; Corpus linguistics; Process (computing); Natural language processing; Psychology; Programming language","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01941919,0.001217938,0.0008561987,0.006440485,0.003478994,0.006543569,0.003473022,0.001172777,0.009553415],"category_scores_gemma":[0.04604601,0.001269513,0.0007868347,0.004009873,0.002676849,0.009421646,0.008548997,0.002347598,0.002692906],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001907728,"about_ca_system_score_gemma":0.004094722,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004127236,"about_ca_topic_score_gemma":0.006554154,"domain_scores_codex":[0.9916826,0.005016739,0.0007991703,0.001065796,0.001155455,0.000280193],"domain_scores_gemma":[0.9295192,0.04773009,0.002205354,0.01118754,0.007777644,0.001580108],"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":[0.0007149841,0.0006790152,0.009415473,0.00301686,0.0001793132,0.001955126,0.1226582,0.008326292,0.04516894,0.07526957,0.06125521,0.671361],"study_design_scores_gemma":[0.0002848307,0.000345225,0.005372848,0.001243431,0.0001517811,0.001198036,0.03712024,0.04950288,0.06703046,0.02279741,0.8146065,0.000346298],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.09201015,0.001204197,0.8302862,0.0024771,0.0006507845,0.002454575,0.005152215,0.03247172,0.03329313],"genre_scores_gemma":[0.1473405,0.0006461055,0.8227816,0.0002384933,0.0001084773,0.002899979,0.009407165,0.006386829,0.01019088],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01941919,"threshold_uncertainty_score":0.1026998,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04584676063739353,"score_gpt":0.2999979899117502,"score_spread":0.2541512292743567,"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."}}