{"id":"W3104309570","doi":"","title":"Nouvelles réflexions sur la lemmatisation des corpus textuels","year":2006,"lang":"fr","type":"preprint","venue":"HAL (Le Centre pour la Communication Scientifique Directe)","topic":"Linguistics and Discourse Analysis","field":"Arts and Humanities","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Canadian Linguistic Association","funders":"","keywords":"Computer science","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02301577,0.003366082,0.002622335,0.007842647,0.006606887,0.02422417,0.003011747,0.006198195,0.0183918],"category_scores_gemma":[0.1098853,0.003292131,0.004329867,0.008748576,0.01702955,0.02048117,0.005582172,0.01724959,0.007021474],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.007667837,"about_ca_system_score_gemma":0.005764885,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03350543,"about_ca_topic_score_gemma":0.02531436,"domain_scores_codex":[0.9337779,0.0424821,0.004002516,0.004464259,0.01448414,0.0007889709],"domain_scores_gemma":[0.8673811,0.1046927,0.002801873,0.01100823,0.01322969,0.0008864434],"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":[0.0006592613,0.0001339353,0.002183584,0.00199457,0.0004081596,0.001510552,0.02163756,0.002678036,0.02978484,0.7931303,0.02866824,0.1172109],"study_design_scores_gemma":[0.0003995426,0.0001515468,0.009493586,0.001707449,0.0004099463,0.00304699,0.01001589,0.02105531,0.04502087,0.2282187,0.6800846,0.0003955619],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.057708,0.01928729,0.7988738,0.04585162,0.005579327,0.0003262895,0.002512275,0.005009943,0.06485149],"genre_scores_gemma":[0.4981089,0.02464295,0.3792274,0.00557503,0.009546726,0.001546366,0.007179001,0.008970932,0.06520277],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.03350543,"threshold_uncertainty_score":0.1217205,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02343732816515759,"score_gpt":0.2373741606876178,"score_spread":0.2139368325224602,"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."}}