{"id":"W1985913097","doi":"10.1109/eict.2014.6777864","title":"Native Language Identification using probabilistic graphical models","year":2014,"lang":"en","type":"article","venue":"2013 International Conference on Electrical Information and Communication Technology (EICT)","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Graphical model; Artificial intelligence; Support vector machine; Probabilistic logic; Task (project management); Feature (linguistics); Principle of maximum entropy; Point (geometry); Natural language processing; Entropy (arrow of time); Identification (biology); Naive Bayes classifier; Machine learning; Pattern recognition (psychology); Mathematics","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":[0.00269233,0.0009261134,0.0007613821,0.003220255,0.0005977218,0.002512467,0.00147664,0.001359909,0.003461956],"category_scores_gemma":[0.0178009,0.0006157784,0.001361735,0.001755971,0.000901857,0.003524844,0.001452177,0.001621894,0.001807309],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001105036,"about_ca_system_score_gemma":0.0009030392,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006637889,"about_ca_topic_score_gemma":0.006084185,"domain_scores_codex":[0.9974597,0.001345587,0.00008815983,0.000499192,0.0004430032,0.0001644004],"domain_scores_gemma":[0.9898432,0.007826376,0.0008372184,0.0006215845,0.0006865606,0.000185113],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004445033,0.0002716185,0.01817201,0.0002476549,0.0002472611,0.0004723194,0.0006542901,0.6250513,0.002999712,0.09484446,0.01285491,0.24374],"study_design_scores_gemma":[0.000008472148,0.00001357546,0.0006329046,0.00001498162,0.00001351446,0.00006849758,0.00002227667,0.9552936,0.0002729378,0.0428382,0.0008082025,0.00001282918],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04294455,0.0004631982,0.9489942,0.001036109,0.00005764202,0.00006672666,0.0007103662,0.002845436,0.002881739],"genre_scores_gemma":[0.8515342,0.0004904906,0.1421841,0.0002962788,0.0001088196,0.0001435563,0.001743646,0.0002483913,0.003250609],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006637889,"threshold_uncertainty_score":0.01423854,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03629424243890863,"score_gpt":0.3031466464180493,"score_spread":0.2668524039791407,"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."}}