Title :
About the use of semaphores to handle contradictory knowledge sources
Author_Institution :
CRIL-CNRS, Univ. D´´Artois, Lens, France
Abstract :
Many approaches to handle inconsistent knowledge require some information to be dropped in order to ensure logical consistency. In this paper, a syntax-based technique that simply weakens the conflicting information by means of semaphores is proposed and analyzed in the context of the fusion of several possibly mutually conflicting knowledge sources. The proposed approach basically adopts a neutral position w.r.t. two knowledge sources in the sense that it does not conduct one source to predominate the other one. However marking a preference for one source and retracting this preference is easy. When the fusion process is iterated, it is shown that semaphores can conduct sources to become actually rejected. Conditions preventing such a drawback are investigated. It is also shown how this problem can be solved when all sources are available simultaneously and how a preference for less-contradicted sources can be realized as well.
Keywords :
formal logic; knowledge based systems; sensor fusion; set theory; artificial intelligence; conflicting information; contradictory knowledge sources; inconsistent knowledge handling; information dropping; information logical consistency; logic-based approaches; logic-based knowledge fusion; mutually conflicting knowledge fusion; semaphores; syntax-based technique; Artificial intelligence; Information analysis; Layout; Lenses;
Conference_Titel :
Tools with Artificial Intelligence, 2003. Proceedings. 15th IEEE International Conference on
Print_ISBN :
0-7695-2038-3
DOI :
10.1109/TAI.2003.1250193