{"id":"W8258861","doi":"10.7589/0090-3558-29.4.596","title":"Multi-class categorization based on cluster analysis and TFIDF","year":2008,"lang":"en","type":"article","venue":"Journal of Wildlife Diseases","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Categorization; tf–idf; Computer science; Class (philosophy); Artificial intelligence; Humanities; Information retrieval; Natural language processing; Philosophy; Physics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005806712,0.001814045,0.001926173,0.01518428,0.002628299,0.002898117,0.002064298,0.001917108,0.005569869],"category_scores_gemma":[0.012888,0.0002792786,0.002385274,0.008587381,0.0008364425,0.001776464,0.001556378,0.001369636,0.003907248],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001621072,"about_ca_system_score_gemma":0.002004391,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01289168,"about_ca_topic_score_gemma":0.008243349,"domain_scores_codex":[0.9949116,0.0006173585,0.0008255302,0.001447361,0.001446746,0.0007515491],"domain_scores_gemma":[0.9925069,0.002785611,0.0003803853,0.0004846412,0.003483948,0.0003584859],"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":[0.001578342,0.0005901412,0.01155202,0.0006828977,0.0003460982,0.0003949816,0.0009947714,0.007704431,0.01320261,0.001981373,0.02625025,0.9347219],"study_design_scores_gemma":[0.0002763375,0.0008978327,0.03950458,0.0002970796,0.0005530148,0.001124421,0.003040885,0.8887106,0.02235324,0.01645946,0.02640031,0.0003823584],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1327511,0.002650919,0.8236553,0.001381334,0.001945842,0.004062674,0.01165352,0.01441515,0.007484162],"genre_scores_gemma":[0.3841248,0.0005649895,0.5903842,0.0001491641,0.0005884575,0.002909854,0.01558412,0.0004930065,0.005201265],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01518428,"threshold_uncertainty_score":0.03070921,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01863920483842632,"score_gpt":0.2532088505944861,"score_spread":0.2345696457560597,"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."}}