{"id":"W4400967584","doi":"","title":"Analyse non supervisée de corpus de documents pour la recherche de sujets : une nouvelle approche basée sur le clustering et la maximisation des traits","year":2024,"lang":"fr","type":"article","venue":"HAL (Le Centre pour la Communication Scientifique Directe)","topic":"Data Mining Algorithms and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Centre National de la Recherche Scientifique; Université du Québec à Montréal","keywords":"Computer science; Cluster analysis; Humanities; Artificial intelligence; Philosophy","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"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":[0.003677962,0.001672222,0.00196728,0.003364266,0.0009241183,0.002400778,0.001900305,0.002360072,0.002253144],"category_scores_gemma":[0.009396832,0.0007283998,0.001733953,0.001681779,0.0009214674,0.001711601,0.001078214,0.001565018,0.002024337],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009601022,"about_ca_system_score_gemma":0.001836651,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009859408,"about_ca_topic_score_gemma":0.01219319,"domain_scores_codex":[0.9963425,0.001033112,0.0002097206,0.001036656,0.00117179,0.0002061945],"domain_scores_gemma":[0.9925584,0.003864616,0.0003607337,0.0008971459,0.002128955,0.0001901434],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001564922,0.0003764016,0.006245531,0.0007052814,0.0006208395,0.0004492528,0.0007465372,0.04375216,0.1759045,0.004460684,0.006572551,0.7586014],"study_design_scores_gemma":[0.0001194239,0.0005584463,0.01449693,0.0001134625,0.0004060719,0.0008737956,0.0004477553,0.8709767,0.08870826,0.008878082,0.01431907,0.0001020536],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.07320765,0.00267452,0.9168451,0.0005900705,0.0001820321,0.0002534674,0.0007973287,0.003434839,0.00201501],"genre_scores_gemma":[0.3756841,0.001483237,0.5971043,0.0003385858,0.000454796,0.000521856,0.003801757,0.001070542,0.01954083],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.009859408,"threshold_uncertainty_score":0.01960403,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1371768219059175,"score_gpt":0.3074384311851966,"score_spread":0.1702616092792791,"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."}}