{"id":"W4294300070","doi":"10.21428/594757db.8b7cd94b","title":"Quantifying French Document Complexity","year":2022,"lang":"en","type":"preprint","venue":"","topic":"Text Readability and Simplification","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Contrast (vision); Task (project management); Natural language processing; Range (aeronautics); Linguistic sequence complexity; Artificial intelligence; Measure (data warehouse); Information retrieval; Linguistics; Data mining","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"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":[0.001790905,0.0006101639,0.0006022545,0.008072607,0.001122882,0.002806556,0.0004288956,0.0006272147,0.00352511],"category_scores_gemma":[0.01994435,0.0001488691,0.0004785011,0.00554428,0.0008150188,0.002223165,0.001178484,0.0005923146,0.0005647509],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002163954,"about_ca_system_score_gemma":0.001405335,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02362634,"about_ca_topic_score_gemma":0.02125745,"domain_scores_codex":[0.9970707,0.0007582387,0.0002282086,0.0006272247,0.001150672,0.00016491],"domain_scores_gemma":[0.9850257,0.007910125,0.001527443,0.001293945,0.003802972,0.0004397971],"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":[0.0007894867,0.0002003797,0.1645073,0.001908845,0.0005463761,0.0007558233,0.006605186,0.05251653,0.04347285,0.04354121,0.02067137,0.6644846],"study_design_scores_gemma":[0.0001074692,0.0005270522,0.5894175,0.0004208708,0.000327852,0.002272395,0.005257592,0.1617088,0.05880036,0.04079247,0.1400423,0.0003253686],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8224779,0.00320752,0.1330131,0.0007903986,0.0001042161,0.0003234661,0.01425876,0.001839661,0.02398497],"genre_scores_gemma":[0.8978694,0.0008067102,0.0790085,0.00008056849,0.0001044469,0.0003563653,0.01708272,0.0002966035,0.004394756],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02362634,"threshold_uncertainty_score":0.04697764,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1522546582994682,"score_gpt":0.346716417992687,"score_spread":0.1944617596932187,"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."}}