{"id":"W2187127363","doi":"","title":"Linguistic Resources for 2012 Knowledge Base Population Evaluations","year":2012,"lang":"en","type":"article","venue":"Theory and applications of categories","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"NIST; Computer science; Annotation; Knowledge base; Selection (genetic algorithm); Population; Information extraction; Resource (disambiguation); Entity linking; Track (disk drive); Information retrieval; Base (topology); Natural language processing; Data science; Artificial intelligence","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"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":[0.02699335,0.001299082,0.001111653,0.009594987,0.00372364,0.005528688,0.003146323,0.002118306,0.03005424],"category_scores_gemma":[0.1084035,0.0008812732,0.001078824,0.00711704,0.0009689669,0.005065999,0.004611211,0.00251852,0.01306123],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005246426,"about_ca_system_score_gemma":0.008656884,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01554727,"about_ca_topic_score_gemma":0.01603324,"domain_scores_codex":[0.9642731,0.01790312,0.003239013,0.001917228,0.01154284,0.001124616],"domain_scores_gemma":[0.9331685,0.02547824,0.001500433,0.006564398,0.03190123,0.001387202],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001949307,0.001610375,0.005956251,0.002026326,0.0001653018,0.0007599157,0.003685774,0.01798078,0.008958279,0.03669365,0.2872868,0.6329272],"study_design_scores_gemma":[0.001136495,0.001041503,0.01457425,0.001356745,0.0004493171,0.0007225171,0.005620081,0.09013504,0.04559671,0.03676828,0.802114,0.0004849444],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"dataset","genre_scores_codex":[0.1450779,0.00292082,0.4023303,0.007162176,0.002069497,0.01564121,0.137793,0.03085898,0.2561462],"genre_scores_gemma":[0.271078,0.0008807682,0.4938522,0.00157673,0.0003063832,0.01886481,0.1782364,0.00524841,0.02995638],"genre_candidate":"dataset","genre_consensus":null,"teacher_disagreement_score":0.03005424,"threshold_uncertainty_score":0.1427562,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01694563726183529,"score_gpt":0.3279130043055564,"score_spread":0.3109673670437211,"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."}}