Title :
Planning responses in reaction to user feedback
Author :
Haller, Susan M. ; Ali, Syed S.
Author_Institution :
Dept. of Comput. Sci. & Eng., Wisconsin Univ., Kenosha, WI, USA
Abstract :
The Interactive Discourse Planner (IDP) plans text to justify and describe domain plans interactively. A speaker uses plan justification to have a listener adopt a recommended plan; he uses plan description to enable a listener to execute a plan. The text plan that IDP formulates and executes incrementally is represented uniformly in the same knowledge base with the domain plans that are under discussion. In this way, the text plan and the domain plans are both accessible for analyzing the listener´s feedback. IDP can interpret vaguely articulated feedback, generate concise replies and metacomments, and detect feedback that initiates digressions. This paper focuses on how IDP plans. We have implemented a system that gives driving directions and route advice interactively. For giving driving route directions and advice, the system does not have a preset agenda. It describes routes and/or justifies a route choice until the listener´s feedback indicates that she is satisfied. Therefore, IDP plans text in reaction to feedback that it uses to try to recognize an expansion of its text plan
Keywords :
expert systems; natural language interfaces; planning (artificial intelligence); Interactive Discourse Planner; domain plans; driving route directions; plan description; plan justification; text planning; user feedback; vaguely articulated feedback; Collaborative work; Computer science; Feedback; System testing; Text recognition;
Conference_Titel :
Tools with Artificial Intelligence, 1995. Proceedings., Seventh International Conference on
Conference_Location :
Herndon, VA
Print_ISBN :
0-8186-7312-5
DOI :
10.1109/TAI.1995.479850