{"id":"W2010177639","doi":"10.1353/ils.2011.0022","title":"Bilingual Document Clustering: Evaluating Cognates as Features / Le groupage de documents bilingues : l’évaluation des cognats comme caractéristiques","year":2011,"lang":"fr","type":"article","venue":"Canadian Journal of Information and Library Science","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Valuation (finance); Linguistics; Cluster analysis; Computer science; Philosophy; Artificial intelligence; Business; Accounting","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":["scholarly_communication"],"consensus_categories":["scholarly_communication"],"category_scores_codex":[0.003200306,0.0002274387,0.0002311764,0.0009818973,0.00087024,0.002566445,0.001235559,0.0001331954,0.0001073167],"category_scores_gemma":[0.001354234,0.0002079943,0.00006533674,0.0009677202,0.00109424,0.0263887,0.000208037,0.0004564713,0.000007016569],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002294094,"about_ca_system_score_gemma":0.01099689,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01410519,"about_ca_topic_score_gemma":0.001308532,"domain_scores_codex":[0.9976661,0.0002054874,0.0007684699,0.0001982505,0.0005730983,0.0005885714],"domain_scores_gemma":[0.9973233,0.0001185065,0.0007451778,0.0002129882,0.000715185,0.0008848017],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00006028709,0.00003398947,0.005523384,0.0003034003,0.00003465405,0.0001511689,0.1584272,0.00009897186,0.0006402784,0.07534032,0.001122315,0.758264],"study_design_scores_gemma":[0.00249494,0.003708952,0.06414717,0.006275945,0.0001566485,0.008204037,0.01752276,0.07112076,0.1356867,0.6781644,0.01052968,0.001987925],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8491631,0.04924849,0.07849869,0.005677109,0.002477637,0.0006998109,0.000031967,0.0001951063,0.01400805],"genre_scores_gemma":[0.7124869,0.0003012197,0.2854349,0.001379248,0.0000729063,0.000002035999,0.000005063761,0.000007985823,0.000309742],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7562761,"threshold_uncertainty_score":0.998469,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05025219599413446,"score_gpt":0.3097175813893523,"score_spread":0.2594653853952179,"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."}}