{"id":"W3016943666","doi":"10.1109/infoteh48170.2020.9066277","title":"Language Distance using Common N-Grams Approach","year":2020,"lang":"en","type":"article","venue":"","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Computer science; Authorship attribution; Similarity (geometry); Natural language processing; Measure (data warehouse); Phenomenon; Artificial intelligence; Stability (learning theory); Data mining; Machine learning","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":[],"consensus_categories":[],"category_scores_codex":[0.000130176,0.00007033195,0.00009789108,0.00001230276,0.00008427806,0.00009443415,0.0004382176,0.00003297718,0.00002033337],"category_scores_gemma":[0.00001855908,0.00006036936,0.00003499797,0.0003261832,0.00001756053,0.0002002148,0.0001454027,0.0001107712,0.00004570313],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001541875,"about_ca_system_score_gemma":0.00001704402,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002149821,"about_ca_topic_score_gemma":0.000001371518,"domain_scores_codex":[0.9993287,0.00005088609,0.000110794,0.0002159719,0.0001338798,0.0001597902],"domain_scores_gemma":[0.9996139,0.00001982318,0.00003422495,0.0002043337,0.00001520449,0.0001124926],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002595013,0.00008431771,0.007892848,0.00009174408,0.0000208615,0.00004472808,0.01776153,0.0012814,0.005697174,0.9343233,0.001303777,0.03147237],"study_design_scores_gemma":[0.0001257999,0.00002010262,0.000212391,0.000003938329,0.000001577966,0.000005533568,0.00027626,0.9911432,0.004740438,0.0003465534,0.002986282,0.0001379853],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01559216,0.00008731824,0.9754891,0.001564387,0.00006480795,0.00006676211,0.000001460653,0.0002815047,0.006852441],"genre_scores_gemma":[0.8768328,6.942541e-7,0.1209486,0.002053948,0.00004717212,0.000001239434,0.000004317568,0.000003713231,0.0001074847],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9898617,"threshold_uncertainty_score":0.2461791,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07502179611688059,"score_gpt":0.3015637118232853,"score_spread":0.2265419157064047,"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."}}