{"id":"W1902658720","doi":"10.5430/air.v4n2p143","title":"A text feature selection method based on category-distribution divergence","year":2015,"lang":"en","type":"article","venue":"Artificial Intelligence Research","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Natural Science Foundation of China","keywords":"Feature selection; Selection (genetic algorithm); Divergence (linguistics); Artificial intelligence; Computer science; Feature (linguistics); Natural language processing; Pattern recognition (psychology); Distribution (mathematics); Mathematics; Statistics; Linguistics; Philosophy","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001310907,0.0008061593,0.001413137,0.00441257,0.0006367632,0.0008860214,0.0007554564,0.0007132524,0.001410614],"category_scores_gemma":[0.003930845,0.0001849258,0.001056453,0.003003627,0.0003559967,0.001198822,0.0005924535,0.0006448157,0.0005851129],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004474888,"about_ca_system_score_gemma":0.0007790976,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002110097,"about_ca_topic_score_gemma":0.001669939,"domain_scores_codex":[0.9985972,0.0001959698,0.0001524018,0.0003190674,0.0006468678,0.00008849129],"domain_scores_gemma":[0.9979739,0.0007173722,0.0001248168,0.0001207236,0.0009919337,0.00007127463],"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":[0.0003496683,0.0002100319,0.007264901,0.0002113208,0.0001419652,0.0002169662,0.0001550203,0.01101736,0.03536054,0.001403556,0.004340441,0.9393281],"study_design_scores_gemma":[0.0002756518,0.0007593982,0.03474488,0.00006719196,0.0002452566,0.001796382,0.0002877053,0.8829726,0.05097982,0.007962786,0.01970541,0.0002028902],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06203923,0.0007558408,0.932987,0.0002298057,0.0001792316,0.0003232782,0.0004950552,0.001713209,0.001277401],"genre_scores_gemma":[0.5301598,0.000382318,0.4628056,0.0001759185,0.0002415391,0.0007574706,0.001988846,0.0001611736,0.003327273],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00441257,"threshold_uncertainty_score":0.006932795,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2404044412835338,"score_gpt":0.445637965919208,"score_spread":0.2052335246356742,"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."}}