{"id":"W2133246278","doi":"10.14778/1453856.1453895","title":"Efficient search for the top-k probable nearest neighbors in uncertain databases","year":2008,"lang":"en","type":"article","venue":"Proceedings of the VLDB Endowment","topic":"Data Management and Algorithms","field":"Computer Science","cited_by":149,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Data mining; Query optimization; Online aggregation; k-nearest neighbors algorithm; Query language; Semantics (computer science); Information retrieval; Sargable; Object (grammar); Point (geometry); Database; Feature (linguistics); Web query classification; Web search query; Search engine; Artificial intelligence","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":[0.003984306,0.0009378948,0.003364607,0.004446898,0.001646222,0.003350858,0.003866679,0.002307604,0.001635337],"category_scores_gemma":[0.02511799,0.001014346,0.0009179089,0.006381348,0.0009277032,0.00732142,0.002427203,0.001219946,0.00058644],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001354977,"about_ca_system_score_gemma":0.001827491,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007188892,"about_ca_topic_score_gemma":0.01182296,"domain_scores_codex":[0.9953053,0.001268094,0.0006558885,0.001016104,0.001424462,0.0003301212],"domain_scores_gemma":[0.9871206,0.009335232,0.0007983377,0.001407169,0.001049508,0.0002892043],"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":[0.001739551,0.0006127961,0.01521179,0.0006535663,0.0003374173,0.0005832411,0.001135874,0.4595737,0.006966825,0.02360196,0.01573184,0.4738514],"study_design_scores_gemma":[0.00005500209,0.0000708286,0.0007603156,0.00001827822,0.00004516274,0.0002457401,0.0003434478,0.9706444,0.002115364,0.0247543,0.0009205624,0.00002666008],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1757375,0.003142703,0.815048,0.00121939,0.00006339722,0.0001918292,0.001227856,0.001691636,0.001677764],"genre_scores_gemma":[0.4844283,0.0006150185,0.5111876,0.0001871843,0.0001036567,0.0001381096,0.00223492,0.0001832989,0.0009218142],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007188892,"threshold_uncertainty_score":0.02107126,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05346318793657699,"score_gpt":0.2715280562480217,"score_spread":0.2180648683114447,"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."}}