Abstract
One of the objectives of the Medoc system is transparent searching in heterogeneous distributed bibliographic and full text data-bases (information providers). For a sufficiently large number of information providers an automatic method for provider selection becomes necessary. We take a decision-theoretic approach: we estimate the cost for retrieving n relevant documents from each provider, then choose the combination which minimizes total costs. The main cost factor is the total number of relevant in each database; additional factors are the retrieval quality of the database, the costs for retrieving a document from the database, and the user-specific costs for retrieving a non-relevant or a relevant document, respectively. In this paper, we also describe a first implementation of the approach where each factor is made explicit so that the implementation can easily be adapted for specific situations.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Fuhr, N. Optimum Database Selection in Networked IR. In Callan, J. and Fuhr, N., editors, NIR'96. Proceedings of the SIGIR'96 Workshop on Networked Information Retrieval, 1996. http://SunSite.Informatik.RWTH-Aachen.DE/Publications/CEUR-WS/Vol-7/.
Gövert, N. Datenbankselektion in vernetzten Information-Retrieval-Systemen. Diplomarbeit, Universität Dortmund, Fachbereich Informatik, April 1997. (http://ls6-www.cs.uni-dortmund.de/~goevert/diploma/).
Gravano, L., Garcia-Molina, H., and Tomasic, A. The Effectiveness of GLOSS for the Text Database Discovery Problem. In Snodgrass, R. T. and Snodgrass, M., Snodgrass, W., editors, Proceedings of the 1994 ACM SIGMOD. International Conference on Management of Data., pages 126–137, New York, 1994. ACM.
Gravano, L. and a Molina, H. G. Generalizing Gloss to Vector-Space Databases and Broker Hierarchies. In Proceedings of the 21st VLDB Conference, 1995.
Nie, J. An Outline of a General Model for Information Retrieval Systems. In Chiaramella, Y., editor, 11th International Conference on Research & Development in Information Retrieval, pages 495–506, Grenoble, France, June 1988. Presses Universitaires de Grenoble.
Robertson, S. The Probability Ranking Principle in IR. Journal of Documentation, 33: 294–304, 1977.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1998 Springer-Verlag
About this chapter
Cite this chapter
Dreger, M., Fuhr, N., Großjohann, K., Lohrum, S. (1998). Provider selection — Design and implementation of the Medoc Broker. In: Barth, A., Breu, M., Endres, A., de Kemp, A. (eds) Digital Libraries in Computer Science: The MeDoc Approach. Lecture Notes in Computer Science, vol 1392. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0052514
Download citation
DOI: https://doi.org/10.1007/BFb0052514
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-64493-4
Online ISBN: 978-3-540-69790-9
eBook Packages: Springer Book Archive