{"id":"W4303437372","doi":"10.7202/1092197ar","title":"Quantitative questions on big data in translation studies","year":2022,"lang":"en","type":"article","venue":"Meta Journal des traducteurs","topic":"Translation Studies and Practices","field":"Arts and Humanities","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Variety (cybernetics); Data science; Big data; Field (mathematics); Natural language processing; Corpus linguistics; Translation (biology); Translation studies; Sentiment analysis; Artificial intelligence; Machine translation; Linguistics; Data mining; Mathematics","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":"codex-gemma-dda1882f352a","candidate_categories":["sts"],"consensus_categories":[],"category_scores_codex":[0.001180551,0.0001368339,0.0002804866,0.0002389508,0.001414262,0.0001560986,0.0002718144,0.000009455823,0.0006117424],"category_scores_gemma":[0.00006615823,0.0001062496,0.00009758249,0.0001352768,0.0002062959,0.0006747644,0.00003498135,0.0004608593,0.00001217061],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004145338,"about_ca_system_score_gemma":0.00003128902,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001070914,"about_ca_topic_score_gemma":0.003141779,"domain_scores_codex":[0.9983761,0.0005426013,0.0003619076,0.0002044641,0.000337548,0.0001773967],"domain_scores_gemma":[0.9989249,0.0005959824,0.000166719,0.0001998688,0.0000749359,0.00003764809],"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":[0.0003799316,0.0004124726,0.0002469785,0.0000402034,0.002556566,0.0000556019,0.05062594,0.001881355,0.00005067516,0.1402572,0.003348342,0.8001447],"study_design_scores_gemma":[0.0003791579,0.0003382445,0.0004021671,0.00001424906,0.0004461957,0.00002600349,0.01027756,0.0000805519,0.000003017928,0.01542472,0.9724711,0.0001370374],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"empirical","genre_scores_codex":[0.09047894,0.8673225,0.001575249,0.0201448,0.006619442,0.000486504,0.0007749646,0.0001007445,0.01249686],"genre_scores_gemma":[0.8876638,0.1073785,0.001814631,0.0007263622,0.001029592,0.00007178768,0.00006861964,0.0000477317,0.001199009],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9691228,"threshold_uncertainty_score":0.9998857,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4622568889228398,"score_gpt":0.3908865321108239,"score_spread":0.07137035681201592,"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."}}