Paper accepted at URSW2008 Workshop (colocated with ISWC2008)
Fuzzy Description Logics have been studied in recent years as a formalism for reasoning under uncertainty in the context of the Semantic Web. Fuzzy Description Logics extend classical DLs and enable DLs to deal with vague knowledge, i.e. concepts and relations which can not be naturally defined in a classical (crisp) way, e.g. all persons that are old. Assertions about the properties of individuals are fuzzy, i.e. they are “true” to a certain extent that is specified by a value in the interval [0;1]. This allows to represent statements such as John is an old person to degree 0.8 or john and marry know each other to degree 0.3. Further, FDLs are para-consistent (to a certain extent): a statement and its negation can hold at the same time without causing an inference system to infer every possible statement about a domain model.
The following paper introduces a new method for performing reasoning with FDLs:
Uwe Keller and Stijn Heymans:
Deciding Fuzzy Description Logics by Type Elimination
The paper received very positive feedback and has been accepted at the 4th International Workshop on Uncertainty Reasoning for the Semantic Web (URSW 2008), October 26, 2008, co-located with the 7th International Semantic Web Conference (ISWC 2008), Karlsruhe, Germany.