Advertisement

A System for Managing Alternate Models in Model-Based Mediation

  • Amarnath Gupta
  • Bertram Ludäscher
  • Maryann E. Martone
  • Xufei Qian
  • Edward Ross
  • Joshua Tran
  • Ilya Zaslavsky
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2405)

Abstract

In [1,3], we have described the problem of model-based mediation (MBM) as an extension of the global-as-view paradigm of information integration. The need for this extension arises in many application domains where the information sources to be integrated not only differ in their export formats, data models, and query capabilities, but have widely different schema with very little overlap in attributes. In scientific applications, the information sources come from different subdisciplines, and despite their poorly overlapping schema, can be integrated because they capture different aspects of the same scientific objects or phenomena, and can be conceptually integrated due to scientific reasons. In the MBM paradigm, a “mediation engineer” consults with domain experts to explicitly model the “glue knowledge” using a set of facts and rules at the mediator. Integrated views are defined in MBM on top of the exported schemas from the information sources together with the glue knowledge source that ties them together. We have successfully applied the MBM technique to develop the KIND mediator or Neuroscience information sources [1]-[4]. To accomplish this, sources in the MBM framework export their conceptual models (CMs), consisting of the logical schema, domain constraints, and object contexts, i.e., formulas that relate their conceptual schema with the global domain knowledge maintained at the mediator. Thus model-based mediation has a hybrid approach to information integration - on the one hand at the mediator integrated views are defined over source CMs and the Knowledge Map using a global-as-view approach; on the other hand, object-contexts of the source are defined as local-as-view.

Keywords

Query Processing Information Integration Knowledge Source Integrate View Query Planning 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    A. Gupta, B. Ludäscher, M. Martone, “Knowledge-based Integration of Neuroscience Data Sources”, 12th Int. Conf. Scientific and Statistical Database Management Systems, Berlin, 39–52, 2000.Google Scholar
  2. 2.
    B. Ludäscher, A. Gupta, M. Martone, “A Mediator System for Model-based Information Integration”, Int. Conf. VLDB, Cairo, 639–42, 2000.Google Scholar
  3. 3.
    B. Ludäscher, A. Gupta, M.E. Martone, “Model-Based Mediation with Domain Maps”, 17th Intl. Conference on Data Engineering (ICDE), Heidelberg, Germany, IEEE Computer Society, 81–90, 2001.CrossRefGoogle Scholar
  4. 4.
    Xufei Qian, Bertram Ludäscher, Maryann E. Martone, Amarnath Gupta: “Navigating Virtual Information Sources with Know-ME”, EDBT 2002: 739–741Google Scholar
  5. 5.
    G. Yang, M. Kifer, “FLORA: Implementing an Efficient DOOD System Using a Tabling Logic Engine”, Computational Logic, 1078–1093, 2000.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Amarnath Gupta
    • 1
  • Bertram Ludäscher
    • 1
  • Maryann E. Martone
    • 1
  • Xufei Qian
    • 1
  • Edward Ross
    • 1
  • Joshua Tran
    • 1
  • Ilya Zaslavsky
    • 1
  1. 1.San Diego Supercomputer CenterUniversity of California San DiegoUSA

Personalised recommendations