{"id":"W1967963033","doi":"10.7202/1006182ar","title":"Dutch Parallel Corpus: A Balanced Copyright-Cleared Parallel Corpus","year":2011,"lang":"en","type":"article","venue":"Meta Journal des traducteurs","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":77,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Natural language processing; Artificial intelligence; Parallel corpora; Sentence; Corpus linguistics; Text corpus; Metadata; Information retrieval; Machine translation; World Wide Web","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":[0.005251894,0.001062571,0.001250229,0.005861925,0.002719857,0.003205048,0.001938892,0.001331576,0.05314226],"category_scores_gemma":[0.02295033,0.0008695832,0.0005955972,0.009653918,0.0014415,0.003028474,0.003466504,0.001795573,0.01803457],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002209835,"about_ca_system_score_gemma":0.005527705,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0217523,"about_ca_topic_score_gemma":0.02109708,"domain_scores_codex":[0.9943743,0.001718659,0.001140219,0.001081041,0.001477372,0.0002084445],"domain_scores_gemma":[0.9830886,0.005454346,0.0009148661,0.002187153,0.007754955,0.00060026],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00111097,0.0003577867,0.003174839,0.006663521,0.0001719568,0.002749316,0.006874791,0.003820449,0.02939308,0.03541341,0.6726194,0.2376505],"study_design_scores_gemma":[0.0003590038,0.0000585154,0.007613836,0.0003810554,0.00008595819,0.001239149,0.001147376,0.002838818,0.007534906,0.00488001,0.9737443,0.0001170254],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.07467075,0.004951212,0.1393739,0.003113846,0.002780161,0.004982148,0.657518,0.004987152,0.1076229],"genre_scores_gemma":[0.09460726,0.001639151,0.1277671,0.0004519094,0.0003824691,0.008776954,0.7344234,0.005200575,0.02675118],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.05314226,"threshold_uncertainty_score":0.1777785,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04397547829965751,"score_gpt":0.265400361353215,"score_spread":0.2214248830535575,"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."}}