{"id":"W2111814513","doi":"10.14778/2535570.2488330","title":"Partitioning and ranking tagged data sources","year":2013,"lang":"en","type":"article","venue":"Proceedings of the VLDB Endowment","topic":"Complex Network Analysis Techniques","field":"Physics and Astronomy","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Ranking (information retrieval); Information retrieval; Categorization; Partition (number theory); Set (abstract data type); Rank (graph theory); Learning to rank; Focus (optics); Data mining; Social media; Data science; World Wide Web; Artificial intelligence; Mathematics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001848689,0.00009149771,0.0001388283,0.0000304439,0.0001373595,0.0001014149,0.0004233802,0.00001134523,0.0002059319],"category_scores_gemma":[0.000008072168,0.00006394753,0.00004390288,0.0001207603,0.00005736652,0.0002390999,0.0006122765,0.0000783382,0.00000377466],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000008146572,"about_ca_system_score_gemma":0.000005246405,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003964949,"about_ca_topic_score_gemma":0.000002071461,"domain_scores_codex":[0.9993232,0.000004495772,0.0001851661,0.0001914589,0.0001479441,0.0001477639],"domain_scores_gemma":[0.9995285,0.00002353945,0.000176167,0.0001703416,0.00007065346,0.00003076946],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000007561238,0.0001349078,0.8457416,0.00005619531,0.000324725,3.393735e-8,0.0008974551,0.00001886999,0.04420003,0.05076668,0.01935875,0.03849314],"study_design_scores_gemma":[0.001979,0.0001338487,0.1291217,0.0008836778,0.0007901069,0.000005463774,0.004948171,0.02630422,0.4293917,0.3756799,0.02961612,0.001146113],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9902109,0.0001132475,0.0003596439,0.0006089166,0.00002392392,0.0003287639,0.000006061445,0.00004177132,0.008306759],"genre_scores_gemma":[0.9978105,0.000005943015,0.001842751,0.00003201535,0.000116623,0.00005012619,0.000005679486,0.00000821709,0.0001281827],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7166199,"threshold_uncertainty_score":0.2607704,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01660096490705265,"score_gpt":0.2322763896575149,"score_spread":0.2156754247504622,"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."}}