Signifying Affordances for Effective Interaction of Agents on the Web

Having declarative specifications of interaction in Web-based Multi-Agent Systems (MAS) could support autonomous agents in discovering and reasoning about affordances11 on the Web towards effectively achieving their goals [2]. One possible path forward is to consider signifiers as a first-class abstraction in Web-based MAS for modelling cues and signs that reveal information about how to exploit affordances offered to agents in hypermedia environments [8]. Signifiers enable agents to interact with shared artifacts and other resources in a hypermedia-driven manner, and recommend under which circumstances interaction should take place based on an agent’s abilities and the agent-environment context. As a result, the exploitability and relevance of an affordance can be dynamically evaluated to guide interaction even in affordance-rich and open environments, for instance, based on the agent’s ability to reason and act using specific methods and mechanisms, the agent’s ability to handle abstractions and processes of a given domain, the agent’s goals or the artifact’s state.