Abstract
This paper introduces a novel extension to the object-oriented RETE algorithm, designed to create networks whose behaviour can be configured by plugging different modules in. The main feature is the possibility of asserting not just new objects as facts, but also information on how the facts satisfy the different constraints in the network. The underlying reasoning process has been created to process imperfect information, for example fuzzy or probabilistic, but the same framework can easily be adapted to reason with defeasible rules, both boolean and imperfect, by choosing the configuration modules appropriately. © 2009 Springer-Verlag Berlin Heidelberg.
Cite
CITATION STYLE
Sottara, D., Mello, P., & Proctor, M. (2009). Towards modelling defeasible reasoning with imperfection in production rule systems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5858 LNCS, pp. 345–352). https://doi.org/10.1007/978-3-642-04985-9_32
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
