{"id":"W2038522479","doi":"10.5555/3191835.3191975","title":"Estimating the size of hidden data sources by queries","year":2014,"lang":"en","type":"article","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Windsor","funders":"","keywords":"Computer science; Variance (accounting); Variety (cybernetics); Data mining; Sample (material); Competitor analysis; Sample size determination; Baseline (sea); Information retrieval; Artificial intelligence; Statistics; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01025272,0.001258673,0.002066756,0.004129885,0.00120933,0.003272018,0.003352352,0.002046494,0.0007535008],"category_scores_gemma":[0.05220307,0.00149047,0.001882112,0.003891463,0.001543056,0.01040824,0.003511744,0.002379622,0.0004191731],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00154825,"about_ca_system_score_gemma":0.00192493,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007056588,"about_ca_topic_score_gemma":0.004575621,"domain_scores_codex":[0.9937676,0.001995188,0.0006777549,0.001767568,0.001411214,0.0003807633],"domain_scores_gemma":[0.9447263,0.04386817,0.00279871,0.005525656,0.002626419,0.0004546535],"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":[0.001768932,0.0006691513,0.1265119,0.002142223,0.001012986,0.001461338,0.004444323,0.2117403,0.05486923,0.05051109,0.01736227,0.5275062],"study_design_scores_gemma":[0.0001170341,0.000121798,0.01398783,0.00009747998,0.0002485927,0.0006612141,0.0007195325,0.9204741,0.0202833,0.03784006,0.00534842,0.0001007246],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1584476,0.003313487,0.8309913,0.001776958,0.00008320284,0.0002990048,0.002495349,0.001317618,0.00127552],"genre_scores_gemma":[0.7517394,0.001830817,0.238899,0.0003558547,0.0003699033,0.0005020811,0.005029156,0.0002295225,0.001044172],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01025272,"threshold_uncertainty_score":0.05422229,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0210675277415002,"score_gpt":0.2535414358595014,"score_spread":0.2324739081180012,"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."}}