A Semiotic - Conceptual Framework for Knowledge Repre - sentation
نویسنده
چکیده
This paper argues that a semiotic-conceptual framework is suitable for knowledge representation because it combines conceptual structures with semiotic aspects. The advantages of such a framework are discussed and explained using an example from an ontology language.
منابع مشابه
Itlich. an Empirical Analysis of Optimization Techniques for Terminological Repre- Sentation Systems. Applied Intelligence, 4(2):109{132, April 1994. Kluwer Academic Publishers. Special Issue on Knowledge Base Management. Edited
متن کامل
An Introduction to Semiotic-Conceptual Analysis with Formal Concept Analysis
This paper presents a formalisation of Peirce’s notion of ‘sign’ using a triadic relation with a functional dependency. The three sign components are then modelled as concepts in lattices which are connected via a semiotic mapping. We call the study of relationships relating to semiotic systems modelled in this manner a semiotic-conceptual analysis. It is argued that semiotic-conceptual analyse...
متن کاملUnderstanding Spontaneous Negotiation Dialogue
In this paper we present the task oriented context representation and the dialogue man ager for the Verbmobil translation system We show how to utilize statistical methods shallow extraction and propositional repre sentation to provide translation relevant in formation and most of all to enable the sys tem to automatically create a dialogue script and result summary
متن کاملA VRML Java Framework for D Objects Streaming over the Internet
In this paper a VRML Java framework for stream ing of D objects over the Internet is presented VRML is a le format for describing D virtual objects and Java is a general purpose program ming language used in a variety of applications They are both powerful and portable languages widely accepted worldwide External Author ing Interface EAI is a set Java classes used for the interaction between th...
متن کاملGyan: a Methodology for Rule Extraction from Artiicial Neural Networks
Arti cial neural network ANN learning methods provide a robust and non linear approach to approximating the target function for many classi ca tion regression and clustering problems ANNs have demonstrated good pre dictive performance in a wide variety of practical problems However there are strong arguments as to why ANNs are not su cient for the general repre sentation of knowledge The argume...
متن کامل