Search

The ability to represent real-world objects is an important feature of a practical knowledge system. Most knowledge systems involve informal or ad-hoc mappings from their internal symbols to objects and concepts in their environment. This work introduces a framework for formally associating symbols to their meanings, a process we call grounding. Two kinds of grounding are discussed with respect to conceptual graphs – active grounding, which involves actors to provide mappings to the environment, and terminological grounding, which involves actors that establish the basic elements of meaning with respect to a subject field΄s agreed-upon terminology. The work incorporates active knowledge systems and international terminological standards. [ABSTRACT FROM AUTHOR]

A software product-line is a set of products built from a core set of software components. Although software engineers develop software product-lines for various application types, they are most commonly used for embedded systems development, where the variability of hardware features requires variability in the supporting firmware. Feature models are used to represent the variability in these software product-lines. Various feature modeling approaches have been proposed, including feature diagrams, domain specific languages, constraint languages, and the semantic web language OWL. This paper explores a conceptual graph approach to feature modeling in an effort to produce feature models that have a more natural, and more easily expressed mapping to the problem domain. It demonstrates the approach using a standard Graph Product-line problem that has been discussed in various software product-line papers. A conceptual graph feature model is developed for the graph product-line and it is compared to other feature models for this product-line. [ABSTRACT FROM AUTHOR]