International Journal of Man-Machine Studies - Special Issue: Knowledge Acquisition for Knowledge-based Systems. Part 4
Relational database theory
Query approximate answering system for an incomplete DKBS
Fundamenta Informaticae - Special issue: intelligent information systems
Boolean Reasoning for Decision Rules Generation
ISMIS '93 Proceedings of the 7th International Symposium on Methodologies for Intelligent Systems
Reducing Information Systems with Uncertain Attributes
ISMIS '96 Proceedings of the 9th International Symposium on Foundations of Intelligent Systems
On the Unknown Attribute Values in Learning from Examples
ISMIS '91 Proceedings of the 6th International Symposium on Methodologies for Intelligent Systems
| Hi-index | 0.00 |
In this paper, we propose a new query answering system for an incomplete Cooperative Knowledge-Based System(CKBS). CKBS is a collection of autonomous knowledge-based systems called agents which are capable of interacting with each other. In the first step of the query processing strategy, the contacted site of CKBS will identify all locally incomplete attributes used in a query. An attribute is locally incomplete if there is an object in a local information system with an incomplete information on this attribute. The values of all locally incomplete attributes are treated as concepts to be learned at other sites of CKBS (see [6]). Rules discovered at all these sites are sent to the site contacted by the user and used locally by the query answering system to replace an incomplete information by values provided by the rules. In the second step of the query processing strategy, an incomplete information is removed from the local information system in a maximal number of places. Next, the query answering system finds the answer to a user query in a usual way (similar to CKBS query answering system).