{"id":"W4245642361","doi":"10.1109/ideas.2007.4318085","title":"An Approach for Text Categorization in Digital Library","year":2007,"lang":"en","type":"article","venue":"","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"","keywords":"Categorization; Computer science; Hierarchy; Text categorization; Information retrieval; Digital library; Document classification; Library classification; Artificial intelligence; World Wide Web","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":[0.00008635662,0.00004873592,0.00004490837,0.0001403273,0.00002755506,0.0002715504,0.0004735745,0.00004741487,0.000005417324],"category_scores_gemma":[0.00001402629,0.00004094139,0.00001374591,0.0003533571,0.00001725398,0.002408947,0.00005463467,0.00003075127,0.000008676444],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001010203,"about_ca_system_score_gemma":0.00001561473,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001215093,"about_ca_topic_score_gemma":5.471217e-7,"domain_scores_codex":[0.9994941,0.000002131253,0.0001221597,0.0001891084,0.00006305851,0.0001294915],"domain_scores_gemma":[0.9996525,0.00002881652,0.00002743697,0.0002545259,0.00001166532,0.00002506764],"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":[0.000002800283,0.00008631114,0.008343033,0.000002944885,7.22135e-7,2.794811e-7,0.00006723114,0.00001744404,0.0001716323,0.7947533,0.0005903449,0.1959639],"study_design_scores_gemma":[0.001448942,0.0003870776,0.07039861,0.00000582119,0.00000241164,0.000006930577,0.001663743,0.4663469,0.0815481,0.3429004,0.03447501,0.0008160058],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005092622,0.000009881316,0.9770643,0.0003477125,0.00002991736,0.0001438279,0.000001093258,0.0006691599,0.01664155],"genre_scores_gemma":[0.8538114,0.000001373626,0.1453113,0.00007220349,0.00001275109,0.00001336637,0.00003693329,0.000003559414,0.0007371104],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8487188,"threshold_uncertainty_score":0.2618567,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01624459706194404,"score_gpt":0.2479245301932853,"score_spread":0.2316799331313413,"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."}}