{"id":"W2970751912","doi":"10.4000/jtei.2301","title":"Encoding Disappearing Characters: The Case of Twentieth-Century Japanese-Canadian Names","year":2019,"lang":"en","type":"article","venue":"Journal of the Text Encoding Initiative","topic":"Digital Humanities and Scholarship","field":"Arts and Humanities","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Glyph (data visualization); Unicode; Kanji; Computer science; Geocoding; Markup language; Variation (astronomy); Cover (algebra); Physiognomy; Font; World Wide Web; Chinese characters; Linguistics; Natural language processing; Artificial intelligence; Geography; Visualization; Cartography; XML; Engineering; Sociology","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":[0.00176423,0.0003494668,0.000266563,0.001091534,0.01044259,0.005275723,0.001200145,0.001516604,0.003694887],"category_scores_gemma":[0.008326568,0.0002986695,0.0002688819,0.003302732,0.008417438,0.003468411,0.002245949,0.00207687,0.0003955727],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0207744,"about_ca_system_score_gemma":0.013666,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.8242042,"about_ca_topic_score_gemma":0.8632104,"domain_scores_codex":[0.9980798,0.000503116,0.00006969426,0.0001560446,0.0006856597,0.0005056023],"domain_scores_gemma":[0.9958663,0.001566145,0.0002720819,0.0005780853,0.001360414,0.0003569922],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"qualitative","study_design_scores_codex":[0.0003382652,0.00005288296,0.01284717,0.0002002645,0.00001280655,0.01635658,0.3661779,0.001892457,0.006926829,0.4909708,0.01134958,0.09287426],"study_design_scores_gemma":[0.00005514301,0.00008273825,0.01964159,0.0003166314,0.0001004084,0.008706708,0.3016186,0.008364712,0.0151385,0.03433929,0.6114256,0.0002100558],"study_design_candidate":"qualitative","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7670706,0.001351163,0.01606533,0.005985222,0.0001943399,0.0000821488,0.0003478623,0.0003358138,0.2085676],"genre_scores_gemma":[0.9605852,0.0004684382,0.006565211,0.0002967237,0.00002198259,0.00001588913,0.0001522496,0.000134634,0.0317596],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1757958,"threshold_uncertainty_score":0.3536621,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05268278347355442,"score_gpt":0.2450025990435659,"score_spread":0.1923198155700115,"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."}}