Agent-Oriented Requirements Analysis from Scenarios

This paper proposes a scenario-driven agent-oriented requirement analysis methodology, where scenarios are textual descriptions of the interactions between various agents described by end-users. These scenarios are then transformed into an internal representation - Scenario-Trees. An inductive learning procedure is designed to decompose, cluster, and generalise the scenario descriptions to obtain an abstract grammar - an attribute grammar. The attributes and attribute computing rules are used to reinforce the expressiveness of the grammar. The various possible types of agents are analysed, the patterns of various agent types are defined. An experimental system SSAS, which generates Agent-Z specification for system from the original scenarios is implemented.