With no doubt the Internet will achieve advantages in exploiting software agents for applications, thanks to their autonomy in carrying out tasks. In such a scenario, appropriated methodologies are needed to manage the interactions among agents. The BRAIN framework proposes an interaction model based on roles, each one composed by a set of capabilities and an expected behavior. The achieved advantages are agent-oriented features, separation of concerns and reuse of solutions and experiences. In this paper we present two interaction infrastructures, Rolesystem and RoleX, which rely on the above mentioned role-based interaction model. These infrastructures allow agents to assume roles and to interact. An application example and the comparison with other approaches show the effectiveness of our approach.
Role-based interaction infrastructures for Internet agents / Cabri, Giacomo; Leonardi, Letizia; Zambonelli, Franco. - In: IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS. - ISSN 0916-8532. - STAMPA. - E86D:(2003), pp. 2262-2270.