{"id":"W2055849934","doi":"10.1111/j.1467-8640.2010.00357.x","title":"AN EFFICIENT MODEL FOR ENHANCING TEXT CATEGORIZATION USING SENTENCE SEMANTICS","year":2010,"lang":"en","type":"article","venue":"Computational Intelligence","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Natural language processing; Categorization; Computer science; Sentence; Phrase; Artificial intelligence; Term (time); Meaning (existential); Semantics (computer science); Set (abstract data type); Word (group theory); Linguistics; Psychology","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":[0.001331001,0.0008535059,0.0009180296,0.001495021,0.0004261447,0.0009968873,0.00155373,0.001009136,0.001903267],"category_scores_gemma":[0.003019782,0.0002955026,0.001099206,0.001235592,0.0004004305,0.003117959,0.0006472496,0.0008648352,0.001424945],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008615638,"about_ca_system_score_gemma":0.001325466,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002626188,"about_ca_topic_score_gemma":0.003002922,"domain_scores_codex":[0.9993637,0.0001525632,0.00006067872,0.0001554116,0.000227983,0.00003961848],"domain_scores_gemma":[0.9989824,0.0004283587,0.00007120575,0.00008491247,0.0004022018,0.00003097868],"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":[0.0004444062,0.0006673452,0.002925491,0.0005928369,0.0002124324,0.0003629664,0.000532882,0.1719816,0.04053913,0.0497116,0.01567585,0.7163535],"study_design_scores_gemma":[0.00001961921,0.00007909678,0.0003334152,0.00001451592,0.0000367769,0.00009696013,0.00002740387,0.9810636,0.003245946,0.01222377,0.002842544,0.00001635795],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01120844,0.0002308819,0.9856244,0.0002604157,0.00008077394,0.0001646997,0.0003003263,0.00134428,0.0007858232],"genre_scores_gemma":[0.2462437,0.0005369655,0.7453707,0.0003366738,0.0002099254,0.00112979,0.001804207,0.0001722247,0.004195841],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002626188,"threshold_uncertainty_score":0.00703907,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05018632138260581,"score_gpt":0.3334549231747679,"score_spread":0.2832686017921621,"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."}}