{"id":"W4303437235","doi":"10.7202/1092191ar","title":"The Multilingual Corpus of Survey Questionnaires: A tool for refining survey translation","year":2022,"lang":"en","type":"article","venue":"Meta Journal des traducteurs","topic":"Interpreting and Communication in Healthcare","field":"Health Professions","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Horizon 2020 Framework Programme; European Commission","keywords":"Comparability; Norwegian; German; Portuguese; Scope (computer science); Linguistics; Catalan; Corpus linguistics; Czech; Computer science; European Portuguese; Natural language processing; European Social Survey; Psychology; Political science; Politics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"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":[0.08390589,0.002192077,0.002124315,0.0221906,0.005258383,0.007375869,0.003387481,0.001671944,0.03648921],"category_scores_gemma":[0.2881936,0.002757189,0.001185435,0.02401361,0.003168064,0.007116633,0.01497157,0.003715831,0.014233],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00464083,"about_ca_system_score_gemma":0.01592527,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008448383,"about_ca_topic_score_gemma":0.008151359,"domain_scores_codex":[0.9009639,0.06363584,0.01778147,0.007644739,0.009169624,0.0008043738],"domain_scores_gemma":[0.6797068,0.1961462,0.01416571,0.05007423,0.0579328,0.001974283],"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":[0.0008749289,0.0005029683,0.008052244,0.009504449,0.0002684892,0.001678335,0.1216674,0.003373007,0.009065305,0.06214694,0.2038286,0.5790372],"study_design_scores_gemma":[0.000410212,0.0001871329,0.01187002,0.003420181,0.000119864,0.0006728512,0.0163351,0.009021002,0.006183729,0.01999772,0.9313962,0.0003858959],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04359079,0.001716431,0.7312778,0.003157499,0.002566464,0.02460928,0.1147859,0.02294476,0.05535107],"genre_scores_gemma":[0.0711263,0.0008439727,0.765449,0.0006728365,0.0002975583,0.05599744,0.08608292,0.0116261,0.007903745],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.08390589,"threshold_uncertainty_score":0.443742,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2630255742535194,"score_gpt":0.4549633573967401,"score_spread":0.1919377831432207,"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."}}