{"id":"W2016325519","doi":"10.1016/j.ipl.2003.09.002","title":"A new differential LSI space-based probabilistic document classifier","year":2003,"lang":"en","type":"article","venue":"Information Processing Letters","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":21,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Northern British Columbia","funders":"","keywords":"Classifier (UML); Probabilistic logic; Probabilistic classification; Computer science; Search engine indexing; Artificial intelligence; Pattern recognition (psychology); Data mining; A priori and a posteriori; Document classification; Adaptability; Machine learning; Support vector machine; Naive Bayes classifier","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":[0.001188324,0.0006479241,0.001552768,0.002576335,0.0008110926,0.001757364,0.00189342,0.001186156,0.00368075],"category_scores_gemma":[0.002225165,0.0003160977,0.0008805097,0.002327062,0.0005291114,0.002772886,0.001532103,0.001087392,0.003641387],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007419487,"about_ca_system_score_gemma":0.001384441,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001874523,"about_ca_topic_score_gemma":0.002566508,"domain_scores_codex":[0.9980991,0.0001944043,0.0001148645,0.0003303493,0.001129663,0.0001316619],"domain_scores_gemma":[0.9988662,0.0002393232,0.0000702782,0.0001779411,0.0005756838,0.00007062799],"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":[0.0003115185,0.0001837218,0.001605414,0.0001292464,0.00006462381,0.00008552875,0.00004543299,0.007016772,0.03513928,0.00658657,0.007404163,0.9414276],"study_design_scores_gemma":[0.00004370655,0.0002139353,0.002131874,0.00001761053,0.00009017247,0.0007700149,0.00003328452,0.9458182,0.03148755,0.005711124,0.01362372,0.00005887488],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01824903,0.001110075,0.9736264,0.0003125898,0.0002875182,0.0001006625,0.0003724873,0.003134225,0.002806987],"genre_scores_gemma":[0.2873691,0.0009865219,0.6909477,0.0005192067,0.0005526026,0.000315207,0.002370823,0.0002674306,0.01667142],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00368075,"threshold_uncertainty_score":0.01231331,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01159156178863901,"score_gpt":0.2222046444828678,"score_spread":0.2106130826942288,"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."}}