Abstract
Suva in cooperation with SYNLOGIC AG developed a knowledge based system for the assignment of companies to risk related groups. The system is fully integrated in the preexisting information infrastructure and almost transparent for the occasional user. The architecture of the system is characterized trough the orthogonal use of a generic model of knowledge and a small set of parametrizeable rule types. The system implements a client server model in which the rule base is held centrally. Rule sets are loaded on demand from the inference engine residing on local PC. The rule based is maintained by domain experts using a specially developed rule editor. The system went operational starting in July 1998 and has since been used routinely. A list of realized benefits is provided.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Lebsanft, E., et al.: Ein wissensbasiertes System für das Versicherungswesen. Informatik- Informatique Magazine (June 1995)
Lebsanft, E.: WEKI calculates premiums for Swiss insurance company. AI Watch magazine (November 1994)
Schreiber, G., Wielinga, B., Breuker, J.: KADS, a principled approach to knowledge-based system development. Academic Press, London (1993)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1999 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Kaufmann, G., Braun, F., Egli, K., Hodel, M. (1999). WEkI A Knowledge Based System of Suva for the Assignment of Companies to Risk Related Groups. In: Puppe, F. (eds) XPS-99: Knowledge-Based Systems. Survey and Future Directions. XPS 1999. Lecture Notes in Computer Science(), vol 1570. Springer, Berlin, Heidelberg. https://doi.org/10.1007/10703016_13
Download citation
DOI: https://doi.org/10.1007/10703016_13
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-65658-6
Online ISBN: 978-3-540-49149-1
eBook Packages: Springer Book Archive