Burghard B. Rieger:
Whereas most cognitive approaches in the study of language have been developing hypotheses concerning the principles of knowing and understanding natural languages (i.e. competence) without bothering too much about communicative language usages in realworld situations (i.e. performance), new semiotic approaches in cognitive computational linguistics explore the procedures believed to underlie processes of language learning and understanding. They do so by simulating these capabilities as system behaviour under recourse to modeled structures, observable in very large samples of situated natural language discourse and represented in vector space formats via numerically specified by quantitative methods of dynamic (re-)construction. - It will be argued that the ecological understanding of informational systems in Computational Semiotics corresponds well to the procedural modeling and numerical reconstruction of processes that simulate the constitution of meanings and the interpretation of signs (semiosis). - The theories of fuzzy sets and possibility distributions together with their derivatives in soft computing appear to be promising in providing suitable formats for computational approaches to natural language processing without the obligation neither to reject nor to accept traditional formal and modeltheoretic concepts or ontologies. Examples from fuzzy linguistic research will be given to illustrate these points.
HTML Format
PDF Format (278 Kb)