Discourse strategies for generating natural-language text
Artificial Intelligence
Principles of artificial intelligence
Principles of artificial intelligence
Tailoring object descriptions to a user's level of expertise
Computational Linguistics - Special issue on user modeling
Generating explanatory discourse
Current research in natural language generation
Generating recipes: an overview of epicure
Current research in natural language generation
Computers and Intractability: A Guide to the Theory of NP-Completeness
Computers and Intractability: A Guide to the Theory of NP-Completeness
A reactive approach to explanation
IJCAI'89 Proceedings of the 11th international joint conference on Artificial intelligence - Volume 2
Evolving questions in text planning
INLG '08 Proceedings of the Fifth International Natural Language Generation Conference
| Hi-index | 0.00 |
Discourse planning systems developed to date apply local considerations in order to generate an initial presentation that achieves a given communicative goal. However, they lack a global criterion for selecting among alternative presentations. In this paper, we cast the problem of planning discourse as an optimization problem, which allows the definition of a global optimization criterion. In particular, we consider two such criteria: (1) generating the most concise discourse, and (2) generating the 'shallowest' discourse, i. e., discourse that requires the least prerequisite information. These criteria are embodied in a discourse planning mechanism which considers the following factors: (1) the effect of a user's inferences from planned utterances on his/her beliefs, (2) the amount of prerequisite information a user requires to understand an utterance, and (3) the amount of information that must be included in referring expressions which identify the concepts mentioned in an utterance. This mechanism is part of a discourse planning system called WISHFULII which generates explanations about concepts in technical domains.