{"id":"W1826301668","doi":"10.1109/ijcnn.2003.1223678","title":"A comparison of SOM based document categorization systems","year":2004,"lang":"en","type":"article","venue":"","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Categorization; Computer science; Artificial intelligence; Self-organizing map; Architecture; Feature (linguistics); Feature vector; Space (punctuation); Unsupervised learning; Text categorization; Machine learning; Data mining; Pattern recognition (psychology); Cluster analysis","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":[0.00275773,0.0005007684,0.0008786174,0.003634532,0.0006860485,0.001903365,0.001161748,0.001014195,0.003798842],"category_scores_gemma":[0.005796335,0.0002326884,0.00057109,0.002646898,0.0002365509,0.002024526,0.0007891395,0.0004011185,0.001937714],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001106223,"about_ca_system_score_gemma":0.0008276453,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004762074,"about_ca_topic_score_gemma":0.00638646,"domain_scores_codex":[0.9983985,0.0002988166,0.0001530615,0.000181841,0.000850872,0.0001169524],"domain_scores_gemma":[0.9962299,0.001195856,0.0000946522,0.0003632641,0.001939024,0.0001772459],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.002486941,0.0006031044,0.009267745,0.001063687,0.0004583788,0.00009999899,0.000282931,0.02117618,0.01641526,0.004097677,0.009349488,0.9346988],"study_design_scores_gemma":[0.0005132349,0.003095888,0.04281564,0.0003559363,0.0006522976,0.0006241755,0.001924787,0.7928196,0.079699,0.009876447,0.06737464,0.0002483931],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.6240831,0.008411445,0.2930193,0.00118031,0.001130554,0.001057525,0.003208958,0.01838568,0.04952323],"genre_scores_gemma":[0.7558211,0.003112702,0.2187496,0.00025323,0.0001960166,0.0004814636,0.005949226,0.0003763,0.01506046],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.004762074,"threshold_uncertainty_score":0.01458442,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03131328734871104,"score_gpt":0.3084639845942755,"score_spread":0.2771506972455645,"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."}}