{"id":"W2159620282","doi":"10.1142/s0218001411009019","title":"IMPACT OF TERM DEPENDENCY AND CLASS IMBALANCE ON THE PERFORMANCE OF FEATURE RANKING METHODS","year":2011,"lang":"en","type":"article","venue":"International Journal of Pattern Recognition and Artificial Intelligence","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Ranking (information retrieval); Feature selection; Computer science; Artificial intelligence; Feature (linguistics); Univariate; Pattern recognition (psychology); Data mining; Curse of dimensionality; Machine learning; Classifier (UML); Ranking SVM; Class (philosophy); Multivariate statistics","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":[0.0162104,0.001318635,0.002318793,0.002160153,0.001287065,0.002028092,0.001215756,0.00132304,0.001308113],"category_scores_gemma":[0.05900905,0.000363915,0.0008065888,0.002154838,0.0006739637,0.002717949,0.001095594,0.001294858,0.00134707],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008533762,"about_ca_system_score_gemma":0.001104199,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003318719,"about_ca_topic_score_gemma":0.002366981,"domain_scores_codex":[0.9894036,0.004287319,0.001024007,0.000972407,0.003618892,0.0006937084],"domain_scores_gemma":[0.9332983,0.05069328,0.002743468,0.006009714,0.006436875,0.0008183342],"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":[0.004184435,0.0009856313,0.02939316,0.0005327411,0.0005190782,0.0003519008,0.0002614808,0.1418463,0.02215008,0.001682115,0.01148842,0.7866046],"study_design_scores_gemma":[0.0002020947,0.001720041,0.03344654,0.00005593118,0.0002057231,0.0006183255,0.000294793,0.9297801,0.02670773,0.003764777,0.003078281,0.0001256776],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7920492,0.01007587,0.1837477,0.001281434,0.0007222374,0.0003489204,0.00106898,0.003459181,0.007246507],"genre_scores_gemma":[0.9494008,0.0007377489,0.04550244,0.0001379911,0.0001781402,0.0001413408,0.001645904,0.0002382263,0.002017549],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0162104,"threshold_uncertainty_score":0.08572978,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1304311421720087,"score_gpt":0.3653728841263023,"score_spread":0.2349417419542936,"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."}}