{"id":"W2949197339","doi":"10.48550/arxiv.cs/0703058","title":"A Comparison of Five Probabilistic View-Size Estimation Techniques in OLAP","year":2007,"lang":"en","type":"preprint","venue":"ArXiv.org","topic":"Data Stream Mining Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Montréal","funders":"","keywords":"Probabilistic logic; Computer science; Reliability (semiconductor); Online analytical processing; Estimation; Hash function; Algorithm; Data mining; Statistics; Mathematics; Artificial intelligence; Data warehouse","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01455552,0.001754056,0.001851931,0.00429491,0.001495495,0.002994427,0.005504038,0.002117654,0.0007753341],"category_scores_gemma":[0.05883826,0.001097008,0.001287483,0.006321726,0.001364893,0.008663634,0.00412123,0.002340131,0.0003251208],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002369672,"about_ca_system_score_gemma":0.00285637,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005598543,"about_ca_topic_score_gemma":0.00433382,"domain_scores_codex":[0.9848887,0.005415585,0.001293068,0.001449331,0.006338526,0.0006148856],"domain_scores_gemma":[0.923286,0.05383174,0.003313478,0.01263934,0.00592582,0.001003627],"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":[0.003432022,0.0008516635,0.02207109,0.001038777,0.0006926557,0.00009167027,0.0006930627,0.1772986,0.008029385,0.01694167,0.005855614,0.7630038],"study_design_scores_gemma":[0.0001387631,0.0007129525,0.00724924,0.00008328083,0.0001724865,0.0002315619,0.000325195,0.9660785,0.009975951,0.01247969,0.002420369,0.0001320863],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2712552,0.007765167,0.7068419,0.0009266372,0.0002235324,0.0004508519,0.0006849295,0.007643478,0.004208277],"genre_scores_gemma":[0.6053646,0.001988879,0.3899229,0.0001309181,0.0001051865,0.0002686481,0.001245705,0.0003787398,0.0005944651],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01455552,"threshold_uncertainty_score":0.07697785,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0650200978567504,"score_gpt":0.3657233901605293,"score_spread":0.300703292303779,"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."}}