{"id":"W2259525462","doi":"","title":"Distance intertextuelle et classement des textes d'après leur structure : méthodes de découpage et analyses arborées","year":2006,"lang":"fr","type":"preprint","venue":"Open Repository and Bibliography (University of Liège)","topic":"Linguistics and Discourse Analysis","field":"Arts and Humanities","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Linguistic Association","funders":"","keywords":"Natural language processing; Computer science; Artificial intelligence; Linguistics; Tree (set theory); Humanities; Mathematics; Combinatorics; Philosophy","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","sts","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0006303051,0.0005339602,0.0008863842,0.004011774,0.001625378,0.001794104,0.001027216,0.0002633138,0.0004631174],"category_scores_gemma":[0.000001883722,0.0005451934,0.0007025797,0.001607236,0.00168044,0.0006760214,0.001229338,0.0005144462,0.000001644931],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000390475,"about_ca_system_score_gemma":0.0001508898,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.1325181,"about_ca_topic_score_gemma":0.01299547,"domain_scores_codex":[0.9973341,0.0006050692,0.0004548843,0.000819472,0.0003728622,0.0004136426],"domain_scores_gemma":[0.9977155,0.0002397066,0.0007505678,0.000586448,0.0004943715,0.000213386],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00034602,0.0009773414,0.6067578,0.001094001,0.003507796,0.0003752062,0.009860025,0.00108409,0.002152702,0.3507469,0.02283462,0.0002635947],"study_design_scores_gemma":[0.002185304,0.0007753436,0.7682217,0.00378496,0.008930827,0.00004121706,0.04378758,0.003227084,0.001178347,0.0651289,0.09991006,0.002828717],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7973364,0.01646895,0.008189823,0.001105476,0.0005492051,0.0005770844,0.001342241,0.00005915308,0.1743716],"genre_scores_gemma":[0.9668074,0.0140506,0.006077464,0.0001173411,0.0002823714,0.00000160927,0.0001684314,0.00002865823,0.01246615],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.285618,"threshold_uncertainty_score":0.9997,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09109063205159147,"score_gpt":0.3248004868262289,"score_spread":0.2337098547746374,"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."}}