Mechanism for Adaptation of Group Decision-making in Multi-agent E-Learning Environment
Intense and stressful group decision-making has become a daily activity in the modern business environments which caused greater interest in systems that allow simulation of group decision-making with agents as human representatives (surrogates). Developm
- PDF / 579,883 Bytes
- 19 Pages / 439.37 x 666.142 pts Page_size
- 105 Downloads / 154 Views
Mechanism for Adaptation of Group Decision-making in Multi-agent E-Learning Environment Denis Mušic´
Abstract Intense and stressful group decision-making has become a daily activity in the modern business environments which caused greater interest in systems that allow simulation of group decision-making with agents as human representatives (surrogates). Development of representative agents is significantly enhanced through use of methods that allow mapping of some of the most important human traits in the world of agents. These traits are emotions, personality and mood which gain importance by their direct effect on the process of individual and therefore group decision-making. In order to provide more stable and efficient group decision-making, this chapter presents the research results of applying concepts of experience and patience to the emotional agents in eLearning environment. Concept of experience is implemented by using Reinforcement learning technique called Q-learning in combination with Self-organizing map, while concept of patience is implemented by introducing a Self-regulation coefficient. Keywords Agents
Patience Q-learning SOM Self-regulation coefficient
9.1 Introduction With the advent of the first movie achievements, the term agent has brought significant amount of mystery, and represented a synonym for an undercover individual that performs various kinds of tasks showing remarkable degree of intelligence and ability. With the primary objective to create more realistic human representatives, intelligence has become the standard trait in the area of software agents which allows them to perform different types of services such as: finding D. Mušic´ (&) Faculty of Information Technologies, University Dzemal Bijedic, Mostar, Bosnia and Herzegovina e-mail: [email protected]
M. Ivanovic´ and L. C. Jain (eds.), E-Learning Paradigms and Applications, Studies in Computational Intelligence 528, DOI: 10.1007/978-3-642-41965-2_9, Springer-Verlag Berlin Heidelberg 2014
251
252
D. Mušic´
the cheapest offer for a specific product, booking airline tickets and hotels, attending meetings, participating in decision-making, choosing best educational material, and etc. Among aforementioned services, particularly interesting is the one in which the agents as human representatives participate in group decisionmaking activities. In almost all working environments such as business and education, individuals show certain limitations when it comes to efficient problem solving because the problems can occur in various forms, and with greater or less degree of complexity. In order to overcome these limitations, it is common to apply the strategy of joining the group where individuals are able to solve parts of problems that match their competencies and expertise. This is supported by extensive research which has shown that the teams perform better than individuals in a broad range of tasks [1]. A strategy of joining is applied to the agent systems and combining with intelligence, emotions, mood and personality
Data Loading...
