Empowering Provenance in Data Integration

  • Authors:

  • Haridimos Kondylakis;Martin Doerr;Dimitris Plexousakis

  • Affiliations:

  • Information Systems Laboratory FORTH-ICS Computer Science Department, University of Crete,;Information Systems Laboratory FORTH-ICS Computer Science Department, University of Crete,;Information Systems Laboratory FORTH-ICS Computer Science Department, University of Crete,

  • Venue:
  • ADBIS '09 Proceedings of the 13th East European Conference on Advances in Databases and Information Systems
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

The provenance of data has recently been recognized as central to the trust one places in data. This paper presents a novel framework in order to empower provenance in a mediator based data integration system. We use a simple mapping language for mapping schema constructs, between an ontology and relational sources, capable to carry provenance information. This language extends the traditional data exchange setting by translating our mapping specifications into source-to-target tuple generating dependencies (s-t tgds). Then we define formally the provenance information we want to retrieve i.e. annotation , source and tuple provenance. We provide three algorithms to retrieve provenance information using information stored on the mappings and the sources. We show the feasibility of our solution and the advantages of our framework.