A Multi Agent Framework (MAFSNUD) for Belief-Desire-Intention (BDI) Model's Decision-Making Problem in Dynamic Situations: An Overview

Intelligent Agents (IA) are a very active area for Artificial Intelligence (Al) research. The Belief, Desire, Intention (BDI) model is a mature and commonly adopted architecture for Intelligent Agents. BDI Agents are autonomous entities able to work in groups and react to changing environmental conditions. However, its architectural design has some problems, which among others, includes its (BDI) agents lacking any specific mechanisms within the architecture to learn from past behavior and adapt to new situations, and the model is only being closely associated with intelligent agents, but does not, of itself, ensure all the characteristics associated with such agents. The question is: how suitable is the BDI model for decision-making of individual intelligent agents in environments which may be dynamic, unpredictable, and uncertain? Does it really possess and exhibit the social ability feature required for intelligent agents' decision-making in dynamic situations? This paper therefore presents an overview of a framework that addresses this issue by synthesizing graded (g)-BDI and Intelligent System (IS) Model paradigms, thus enabling MAFSNUD (Multi-Agent Framework for Social Network Users' Decision-Making) agents suitable in dynamic, unpredictable and uncertain situations, with UTAR's UniCAT (University Cyber Agent Technology) as a case study.