{"id":"W2077512673","doi":"10.1214/07-ba209","title":"Improving classification when a class hierarchy is available using a hierarchy-based prior","year":2007,"lang":"en","type":"article","venue":"Bayesian Analysis","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":33,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Hierarchy; Multinomial logistic regression; Class hierarchy; Computer science; Class (philosophy); Machine learning; Artificial intelligence; Multinomial distribution; Bayesian probability; Tree (set theory); Data mining; Mathematics; Statistics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.005549299,0.001144436,0.001703351,0.003321329,0.001146953,0.00198854,0.002189211,0.00191499,0.003505332],"category_scores_gemma":[0.02292203,0.0008976844,0.001612213,0.002541037,0.001185026,0.00606089,0.002414818,0.004192213,0.002173964],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001581483,"about_ca_system_score_gemma":0.001946552,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01353026,"about_ca_topic_score_gemma":0.01987591,"domain_scores_codex":[0.9958107,0.001294628,0.0001961766,0.0008479599,0.001547048,0.0003034296],"domain_scores_gemma":[0.9855419,0.009237544,0.0008422449,0.002480401,0.001523305,0.0003745367],"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":[0.0004801848,0.0005965664,0.01262573,0.0002473934,0.0002444366,0.0001127593,0.0008138675,0.1058982,0.01087823,0.02547402,0.01493039,0.8276982],"study_design_scores_gemma":[0.00004302853,0.00007484404,0.002860883,0.00004952322,0.00005930132,0.00008897795,0.0000800624,0.9521767,0.003221023,0.03763674,0.003667615,0.00004143557],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02513021,0.0004353309,0.9697622,0.0006038241,0.00004494441,0.00007052,0.0002442473,0.001701898,0.002006726],"genre_scores_gemma":[0.3425012,0.0004245429,0.6512994,0.0004625507,0.0002209419,0.0001943356,0.001284802,0.0003733022,0.003238867],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01353026,"threshold_uncertainty_score":0.02934784,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03222386111863169,"score_gpt":0.2757000272695841,"score_spread":0.2434761661509524,"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."}}