{"id":"W2207847180","doi":"10.14778/2824032.2824049","title":"Fuzzy joins in MapReduce","year":2015,"lang":"en","type":"article","venue":"Proceedings of the VLDB Endowment","topic":"Data Mining Algorithms and Applications","field":"Computer Science","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Victoria","funders":"","keywords":"Joins; Skyline; Computer science; Point (geometry); Hamming distance; Fuzzy logic; Binary number; Algorithm; Data mining; Theoretical computer science; Artificial intelligence; Mathematics; Programming language; Arithmetic","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":[0.0003027384,0.00007274306,0.00009338848,0.00005465074,0.00003416817,0.00006393428,0.00113239,0.00002038411,9.230242e-7],"category_scores_gemma":[0.00005085888,0.00005122048,0.0000308689,0.0004146097,0.00003200295,0.000278739,0.0005297822,0.00007756847,0.00001561221],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000565259,"about_ca_system_score_gemma":0.00004234377,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007825258,"about_ca_topic_score_gemma":0.000002460861,"domain_scores_codex":[0.9992173,0.000002701593,0.0001747326,0.0002024203,0.0002444378,0.0001584053],"domain_scores_gemma":[0.9995451,0.0000103544,0.00009902136,0.0001924584,0.00008959329,0.0000634533],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00001032035,0.000652384,0.007344702,0.00006058307,0.00002955042,0.000001050386,0.006605655,0.00006161057,0.02638866,0.7881371,0.08174744,0.08896091],"study_design_scores_gemma":[0.004182641,0.0004829688,0.04090056,0.0005033286,0.00004381186,0.0001068697,0.002836065,0.03739049,0.4105994,0.3544609,0.1473643,0.001128662],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7318161,0.0006357042,0.00595455,0.03131737,0.001308468,0.001798917,0.00003634868,0.0003718601,0.2267607],"genre_scores_gemma":[0.9404029,0.00001298144,0.05876947,0.0001620389,0.00004676996,0.00008403352,6.187489e-7,0.000006055796,0.0005151261],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4336762,"threshold_uncertainty_score":0.2104282,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03116395072858845,"score_gpt":0.245206021545165,"score_spread":0.2140420708165766,"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."}}