Abstract
Interface agents are one of the most relevant applications of agent technology to assist humans in using computer software. However, the development of agents assisting users working on the web, using multiple applications to browse information, represents a challenging task. Particularly, when these agents need to act on standard web applications whose source code is not available, developers have one additional requirement: the complete independence between the web application and the agent. This paper proposes a method for connecting these two independent components based on a software architecture that specifies parts and relations of this particular type of web agents.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
