Abstract
Fuzzy rule interpolation is an important technique for
performing inferences with sparse rule bases. Even when given observations
have no overlap with the antecedent values of any rule, fuzzy rule
interpolation may still derive a conclusion. Nevertheless, fuzzy rule
interpolation can only handle fuzziness but not roughness. Rough set
theory is a useful tool to deal with incomplete knowledge, which handles
roughness but not fuzziness. Fuzzy rough sets are used to extend the
original concepts in rough sets. This paper proposes a novel rule
interpolation method which integrates fuzzy-rough representations with
rule interpolation to deal with both fuzziness and roughness. The method
follows the approach of [1],[2], using transformation-based techniques to
perform interpolation, and can deal with rule interpolation in a more
flexible and more robust way.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 11th UK Workshop on Computational Intelligence (UKCI2011) |
| Publisher | Manchester University Press |
| Pages | 55-60 |
| Publication status | Published - 07 Sept 2011 |