{"id":"W2324256469","doi":"10.14195/2182-8830_4-1_3","title":"Sentence-Alignment and Application of Russian-German Multi-Target Parallel Corpora for Linguistic Analysis and Literary Studies","year":2015,"lang":"en","type":"article","venue":"Matlit Revista do Programa de Doutoramento em Materialidades da Literatura","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Université de Montréal","keywords":"Computer science; German; Natural language processing; Sentence; Rule-based machine translation; Artificial intelligence; Set (abstract data type); Linguistics; Corpus linguistics; Parsing; Programming language","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0009227116,0.0003715632,0.0006831245,0.0002904969,0.0001123988,0.1093323,0.0005331141,0.0001354007,0.000001255649],"category_scores_gemma":[0.0001463291,0.0003049899,0.0001055175,0.0006098821,0.0001604426,0.0004571934,0.0004115597,0.0001106068,9.300593e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001349919,"about_ca_system_score_gemma":0.00004361555,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002722357,"about_ca_topic_score_gemma":2.437826e-8,"domain_scores_codex":[0.9976911,0.000164199,0.00071558,0.0007183583,0.0003147973,0.0003959752],"domain_scores_gemma":[0.9981168,0.00006700405,0.0005569782,0.0006523203,0.0003859948,0.0002208979],"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":[0.0001756219,0.0002209283,0.007146765,0.002058569,0.0005362643,0.00003204801,0.01986364,0.000002243484,0.004415374,0.9410151,0.00006495076,0.02446848],"study_design_scores_gemma":[0.005499535,0.001734054,0.0002023491,0.002388434,0.001865045,0.0002646993,0.001387843,0.06416618,0.003659302,0.4173728,0.4981948,0.003265008],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4408142,0.04471599,0.511024,0.0003658481,0.0002189497,0.002106892,0.0001513718,0.0005980085,0.000004688467],"genre_scores_gemma":[0.5392437,0.00000504669,0.4602253,0.00007893007,0.00005082786,0.000251757,0.0001038877,0.000016374,0.00002413147],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5236423,"threshold_uncertainty_score":0.9999402,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02821082619746977,"score_gpt":0.3351590564022613,"score_spread":0.3069482302047916,"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."}}