An Integrated Trust Model Enhancing Generic Nodal Abstraction for Modeling Human-Agent Collaboration

• Main Authors: Mohammed, Khudhair Abbas, Mohammed, Muamer N., Mazlina, Abdul Majid, Mohd Sharifuddin, Ahmad
• Format: Article
• Language: English
• Published: American Scientific Publisher 2018
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://umpir.ump.edu.my/id/eprint/19962/1/An%20Integrated%20Trust%20Model%20Enhancing%20Generic.pdf; http://umpir.ump.edu.my/id/eprint/19962/; https://doi.org/10.1166/asl.2018.12962

In most dynamic environment where human and agents are coexisting, agents overwhelming demands to cooperate with inner and outer services to achieve their goals. Along with line, agents are potentially being attracted by deceptive agents. Such situation is definitely not mitigated that risks the entire model and threaten its long-term performances. Due to the uncertainty of an agent and its potential behavior especially in human-agent collaboration systems, an agent might behave fraudulently to its partners (humans or agents). Therefore, trust and reputation merits are essentially determined to be applied in multi-agent systems. Consequently, an agent has to be qualified and constrained with certain criteria that render it trustworthy and reliable. During this process, these criteria are strictly formulated based on the number of tasks and interactions among agents. Over the past decade, many researchers propose various trust models with outstanding features. However, balancing between the enormous features of the proposed trust models, but still there is a lack of research effort in many security aspects which considered a critical challenge. In this paper, we present a trust model for a Generic Nodal Abstraction (GNA) approach that considers the reliability of agents and introduce their adaptation abilities to improve their preferences in systems. This paper gives an overview of some existing trust models in MASs.