Probabilistic Message Passing in Peer Data Management Systems Philippe Cudré-Mauroux, Karl Aberer EPFL Andras Feher, T.U. Darmstadt
Peer Data Management Systems Pairwise mappings –Query reformulation between two peers (schemas) Local mappings overcome global heterogeneity –Transitive closures on mapping operations T18:49:03Z T20:09:28Z date? 05/08/2004 Jan 1, 2005 article weather es:cDate xap:CreateDate es:cDate myRDF :Date myRDF: Date xap:ModifyDate
Problem: Precision/Recall Tradeoff Semantic Query routing –To whom shall I forward a query posed against my local schema? Some (most) mappings will be (partially) faulty –Low expressive power of mappings samePropertyAs / sameClassAs subclassOf –Automatic schema alignment techniques –Different views on conceptualizations Local query resolution –Low recall Flooding (PDMS so far) –Low precision Standard deductive integration is not sufficient –Uncertainty on mappings and conceptualizations
Probabilistic Message Passing Link-based analysis of the PDMS - Automatically deriving quality measures for the mappings Decentralized computations - Analysis of loops and parallel paths - Iterative sum-product computations (loopy belief propagation) a priori information on mapping
Implemented System Schemas –Import from OWL (Web Ontology Language) Mappings –KnowledgeWeb Ontology Alignment API –Import from RDF/XML –Automated on-the-fly creation –Comparison to standard alignments Automatic derivation of quality measures P(m=correct | {F}) for the mappings using iterative message-passing Reference Probabilistic Message-Passing in Peer-Data Management Systems Philippe Cudré-Mauroux, Karl Aberer, and Andras Feher International Conference on Data Engineering (ICDE 06).