Personalization has been identified as a key task to the success of many modern systems. As Riecken (2000, p. 28) writes in the editorial of the special issue of the Communications of the ACM devoted to this subject, “personalization means something different to everyone.” Personalization can take various forms; however, it can be broadly described as the set of mechanisms by which an application is tailored to a particular end user and his or her goal.
This paper presents the development of an agent-based Virtual Private Network (VPN) system that supports multimedia service brokering. The VPN agents employ multi-attribute preference models in order to represent the end-user preferences, and a multi-criteria decision making model to evaluate available services from network providers. A prototype multi-agent system demonstrating the proposed approach has also been implemented.
Boiled yam end‐user preferences and implications for trait evaluation