{"id":"W4297878801","doi":"10.1590/scielopreprints.4755","title":"ARCHIVING AND LANGUAGE DOCUMENTATION","year":2022,"lang":"en","type":"preprint","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Documentation; Relevance (law); Focus (optics); Indigenous; Indigenous language; Library science; Linguistics; Political science; World Wide Web; Computer science","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.008750267,0.0005630715,0.00059461,0.005377841,0.00458478,0.01769144,0.001498368,0.001960515,0.01398318],"category_scores_gemma":[0.02919384,0.0005293687,0.0006512645,0.007011344,0.008135621,0.01462884,0.007953698,0.00328561,0.004546364],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00336208,"about_ca_system_score_gemma":0.006725601,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008002601,"about_ca_topic_score_gemma":0.007093879,"domain_scores_codex":[0.9870507,0.006535875,0.001228051,0.001418859,0.003386351,0.0003801972],"domain_scores_gemma":[0.979788,0.007281359,0.001558857,0.007120811,0.003716763,0.0005341996],"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":[0.00003708526,0.00003285347,0.001273053,0.000631445,0.00002519794,0.0002910718,0.0125339,0.0005540265,0.001918221,0.6347293,0.03428587,0.3136881],"study_design_scores_gemma":[0.000009826814,0.00002472285,0.00118334,0.001028113,0.00002393868,0.001299664,0.005398415,0.00152927,0.002335974,0.2328341,0.7542882,0.00004455961],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"other","genre_scores_codex":[0.0245214,0.04370256,0.4625421,0.05692678,0.004990481,0.0003984501,0.001706787,0.003739416,0.401472],"genre_scores_gemma":[0.3471783,0.04519582,0.435408,0.006505694,0.003452992,0.0004431121,0.00292386,0.002892743,0.1559995],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.01769144,"threshold_uncertainty_score":0.04677844,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01005173041471197,"score_gpt":0.3036480371038187,"score_spread":0.2935963066891067,"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."}}