{"id":"W2010916971","doi":"10.1145/1132863.1132868","title":"Integrating XML data sources using approximate joins","year":2006,"lang":"en","type":"article","venue":"ACM Transactions on Database Systems","topic":"Data Management and Algorithms","field":"Computer Science","cited_by":28,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Joins; XML validation; XML; Efficient XML Interchange; Document Structure Description; XML Schema (W3C); XML database; Information retrieval; Streaming XML; Data mining; Set (abstract data type); Tree (set theory); XML Encryption; Database; Programming language; World Wide Web","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":[0.01359359,0.001268061,0.00234265,0.004707168,0.001813167,0.007316389,0.003677216,0.001997215,0.001168524],"category_scores_gemma":[0.04275217,0.001227675,0.002008241,0.01020504,0.001651267,0.01106493,0.00909385,0.002113866,0.0008928873],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001571089,"about_ca_system_score_gemma":0.002551043,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003814023,"about_ca_topic_score_gemma":0.003764135,"domain_scores_codex":[0.9802822,0.006270072,0.001510047,0.003074972,0.00837563,0.0004871135],"domain_scores_gemma":[0.977114,0.01119548,0.002524092,0.006072334,0.002619655,0.000474393],"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":[0.000540636,0.0002671052,0.01320487,0.0004129099,0.0004222191,0.0007013024,0.002189641,0.3480946,0.01541524,0.1503491,0.004384737,0.4640176],"study_design_scores_gemma":[0.0000320051,0.0001460049,0.0008764026,0.0000440965,0.00006431135,0.0003683513,0.0004017719,0.903037,0.008182772,0.08020989,0.006586715,0.00005068056],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01470884,0.0003687578,0.9829029,0.0001556621,0.00002253231,0.00009485552,0.0001553124,0.0009198168,0.0006712509],"genre_scores_gemma":[0.1569842,0.0004293297,0.8401152,0.0001206393,0.00006784721,0.0001603939,0.001103348,0.0002200469,0.0007988826],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01359359,"threshold_uncertainty_score":0.07189059,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06748389564404962,"score_gpt":0.2828255905583581,"score_spread":0.2153416949143085,"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."}}