A Hybrid Diagnostic-Recommendation System for Agent Execution Applied to Ubiquitous Computing Systems

Andrew D. CostaCarlos J. P. LucenaViviane T. SilvaDonald CowanPaulo AlencarBaldoino Fonseca

When agents of software based systems operate in some domains such as ubiquitous computing, they may not be able to attain their goals owing to failures during system execution. When an agent tries to achieve its desired goals, but faces failures during execution, it becomes important to understand why such failures occurred and what can be done to remedy the problem. Sometimes, the problem relates to the information exchanged among the software agents, which may depend on their levels of trust and reputation. In this paper, we discuss solutions to the main challenges of creating diagnoses and provide recommendations about agent execution to support eventual attainment of goals in ubiquitous computing systems. We also propose a hybrid diagnostic recommendation framework that provides support for different methods of addressing such challenges and we demonstrate how the framework can be used in ubiquitous computing applications.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: