CISC 885: Discourse and Dialogue

Catalog Description:
Pragmatic issues involved in developing intelligent discourse and dialogue systems. Topics: speech acts, characteristics of a coherent dialogue, explicit versus implicit communication, discourse models, planning and plan recognition, ill-formed input, cooperative responses, user models and default reasoning.

Current Texts: Selected papers from journals and conference proceedings .

Goals:
This course concentrates on pragmatic issues involved in developing intelligent discourse and dialogue systems. The emphasis is on leearning fundamental concepts, understanding the major problems in developing robust natural language systems, identifying the merits and deficiencies of proposed solutions to these problems, and creatively proposing new and profitable avenues of research. The overall goal of the course is to prepare the student to pursue research in natural language systems.

Content:

• Utterances, meaning, and intention
• Mutual beliefs and the cooperative principle
• Conversational coherency and models of discourse
• Computational approaches to speech act recognition
• The planning/plan-recognition paradigm
• Approaches to handling ill-formed input
• Cooperation and collaboration in dialogue
• Constructing and reasoning with user models
• Default reasoning in language understanding

Typical Course Requirements: Critical analysis of literature papers, project/term paper.

Required Background: CISC 681 Artifical Intelligence

Helpful Background: CISC 882 Natural Language Processing; A knowledge of first-order predicate logic