{"id":"W2177050153","doi":"","title":"Feature Space Selection and Combination for Native Language Identification","year":2013,"lang":"en","type":"article","venue":"NPARC","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Classifier (UML); Artificial intelligence; Exploit; Feature selection; Spelling; Feature vector; Natural language processing; Support vector machine; Variety (cybernetics); Feature (linguistics); Machine learning; Linguistics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002102737,0.00004518747,0.00004415497,0.00004699398,0.0001195969,0.0001304025,0.0000913383,0.00004574361,0.00001516756],"category_scores_gemma":[0.00008368844,0.00004291898,0.00001402419,0.0001617572,0.000009586071,0.0004011797,0.00002288642,0.00006138901,0.00002026731],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002571434,"about_ca_system_score_gemma":0.0000117107,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000005228038,"about_ca_topic_score_gemma":0.000003026592,"domain_scores_codex":[0.999592,0.00003739758,0.00005262711,0.0001488837,0.00008140062,0.00008767229],"domain_scores_gemma":[0.9996458,0.00005772254,0.00005283993,0.000078001,0.0001351211,0.00003048584],"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":[0.000003943299,0.00002594717,0.000383106,0.00002659674,0.000008784958,1.586127e-7,0.003483943,0.000004690778,0.232147,0.6948307,0.007800466,0.06128467],"study_design_scores_gemma":[0.0007209996,0.0001211519,0.02616435,0.00001923266,0.000009253269,0.00001054646,0.0005313722,0.4864647,0.3080105,0.1737939,0.003897454,0.0002564771],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07660003,0.00003261563,0.9094782,0.01227053,0.0001504661,0.0003670893,0.00000200865,0.0001050817,0.0009939207],"genre_scores_gemma":[0.9723087,0.000002346465,0.02557675,0.00009213941,0.00002716537,0.00005579849,0.00001185878,0.000002886107,0.001922381],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8957086,"threshold_uncertainty_score":0.1750185,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01409262470222885,"score_gpt":0.2742992865545968,"score_spread":0.260206661852368,"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."}}