A semiotic-inspired machine for personalized multi-criteria intelligent decision support

Abstract

The need for appropriate decisions to tackle complex problems increases every day. Selecting destinations for vacation, comparing and optimizing resources to create valuable products, or purchasing a suitable car are just a few examples of puzzling situations in which there is no standard form to find an appropriate solution. Such scenarios become arduous when the number of possibilities, restrictions, and factors affecting the decision rise, thereby turning decision makers into almost mere spectators. In such circumstances, decision support systems (DSS) can play an important role in guiding people and organizations towards more accurate decision making. However, conventional DSS lack the necessary adaptability to account for dynamic changes and are frequently inadequate to tackle the subjectivity inherent in decision-maker's preferences and intention. We argue that these shortcomings can be addressed by a suitable combination of Semiotic Theory and Computational Intelligence algorithms, which together can make up a new generation of DSS. In this article, a formal description of an Intelligent Semiotic Machine is provided and tried out in practical decision contexts. The results obtained show that our approach can provide well-suited decisions based on user preferences, achieving appropriateness while fanning out subjective options without losing decision context, objectivity, or accuracy.