Semiotics and Intelligent Systems Development: An Introduction

Sometimes, in order to mature, a methodology operates outside the mainstream for a period of time; it becomes the mainstream methodology only after this phase. This happened, for example, with neural networks, which appeared in the early 60's, but became the mainstream in research only 25 years later, after the development of the backpropagation algorithm as its learning algorithm. It happened also with fuzzy systems, which became popular only after the appearance of industrial fuzzy control applications in Japan. Likewise, an alternative approach is flourishing outside the mainstream, yet remain unknown to most researchers in AI (Artificial Intelligence) and IS (Intelligent Systems). This alternative approach is Computational Semiotics – and we argue that it is a rapidly evolving and maturing research field that could become the basic operational theory and method within AI and IS. Semiotics is a field of research involved in the study of meaning processes and communication practices within the fields of the natural and social sciences, linguistics and philosophy. Ideas and concepts from semiotics are increasingly being used by different researchers in computer science as a source of both theoretical insights and practical methodology. In particular, we consider that the field of artificial intelligence (intelligent systems) could significantly benefit from the use of semiotic insights and methods. The interdisciplinary method of semiotics, applied to the investigation of sign processes and dedicated to the development of intelligent systems, is often referred to by the scientific community as Computational Semiotics. This approach proposes a new analysis and methodology to the study of intelligent control and intelligent systems, an approach based around an explicit account of the notion of the sign. This strategy has introduced a wealth of both theoretical and methodological tactics developed under the scope of semiotics, that are being used to enrich artificial intelligence and enable it to establish new frontiers and bridge the theoretical and methodological gaps that have disturbed artificial intelligence studies for quite some time. Early attempts of interdisciplinary studies involving semiotics and intelligent systems were independently developed by researchers from Russia and the United States during the 60's and 70's. The original coverage of intelligent control theory by the Russian Dmitri Pospelov (Pospelov et.al. 1977; Pospelov 1991) is still almost completely unknown in western science. In the United States, a similar effort, also unknown to the mainstream, appeared in the work of Eugene Pendergraft (1993). Despite being ignored for about twenty years, a new and growing interest in such an approach began to appear in the 1990's. In 1990, James Fetzer proposed using Peirce's philosophy of the sign as a strategy to deal with traditional problems in AI (Fetzer 1990). In