{"id":"W2260484439","doi":"10.14778/2824032.2824036","title":"SEMA-JOIN","year":2015,"lang":"en","type":"article","venue":"Proceedings of the VLDB Endowment","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":31,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Joins; Join (topology); Hash join; Table (database); Theoretical computer science; Data mining; Database; Programming language; 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":[0.004506487,0.0001046318,0.0001996424,0.0001062757,0.00007158492,0.0001733639,0.001608101,0.00002890498,0.00009238273],"category_scores_gemma":[0.00217122,0.00005761459,0.0001057427,0.0005442548,0.00009932725,0.000371512,0.001061134,0.0000789933,0.0002698129],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005517833,"about_ca_system_score_gemma":0.00003102954,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006553871,"about_ca_topic_score_gemma":0.000007710871,"domain_scores_codex":[0.9971752,0.00001933322,0.0004998872,0.0002998897,0.001802011,0.0002036429],"domain_scores_gemma":[0.9987275,0.00009267046,0.0003451281,0.0003104226,0.0004077383,0.0001165503],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00004749574,0.0001803396,0.004232143,0.00002544698,0.00003692829,4.795619e-7,0.002107809,0.00002234002,0.002424418,0.1956585,0.7783797,0.0168844],"study_design_scores_gemma":[0.0007256248,0.0001126073,0.00270726,0.0000387251,0.00002968387,0.000005298719,0.007430501,0.0001606574,0.03601578,0.2649202,0.68769,0.0001636058],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.4399451,0.000279129,0.0005336356,0.03031991,0.001955734,0.001293543,0.00005282796,0.0001189117,0.5255012],"genre_scores_gemma":[0.9859812,0.000009056158,0.001392154,0.0008535985,0.00007764292,0.00002465139,7.039278e-7,0.00000713933,0.01165383],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5460361,"threshold_uncertainty_score":0.3467989,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2932342444414948,"score_gpt":0.4016846612147633,"score_spread":0.1084504167732685,"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."}}