{"id":"W2051349122","doi":"10.3115/1118853.1118874","title":"Letter level learning for language independent diacritics restoration","year":2002,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":35,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Romanian; Czech; Computer science; Natural language processing; Artificial intelligence; Linguistics","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":[0.001187723,0.000937952,0.0009228521,0.001525094,0.0008373213,0.001427526,0.001975632,0.001255727,0.004652265],"category_scores_gemma":[0.003488105,0.0003949463,0.0006876073,0.0009067425,0.0009528002,0.002284965,0.001420946,0.002281442,0.005411844],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005680123,"about_ca_system_score_gemma":0.00102216,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001105872,"about_ca_topic_score_gemma":0.001704648,"domain_scores_codex":[0.9991898,0.000142986,0.00005541812,0.0003018399,0.0002273259,0.00008273515],"domain_scores_gemma":[0.9980566,0.000654636,0.0001854262,0.0004856252,0.0005221628,0.00009559053],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002975692,0.0001587199,0.0009281401,0.0002182245,0.00004603515,0.0001237413,0.0001831167,0.01127258,0.06062815,0.006643713,0.005352852,0.914147],"study_design_scores_gemma":[0.00007496425,0.0001933509,0.001409845,0.00004828175,0.00006447948,0.0004781659,0.0001987305,0.7881259,0.1502796,0.0274714,0.0315766,0.0000786591],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02109287,0.0003221262,0.9634866,0.0001378124,0.0002273993,0.00006923662,0.0001205092,0.01195092,0.002592554],"genre_scores_gemma":[0.2803091,0.0002533862,0.7044407,0.0002315106,0.000170751,0.0001049023,0.0007267995,0.000901279,0.01286152],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004652265,"threshold_uncertainty_score":0.01556337,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03785590327783317,"score_gpt":0.288233978363281,"score_spread":0.2503780750854478,"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."}}