Plan-based text generation in an on-line help system
This paper describes an on-line help system, ASSIST, which is designed to produce multiparagragh explanations in English in response to questions about computer terminology. ASSIST's text generation mechanism is based on the speech act planning theory proposed by Cohen. Namely, generation process can be viewed as the process of planning utterances to achieve a discourse goal. In order for the system to have the ability to provide different responses in accordance with user's level of expertise, a user model is constructed throughout a session, and then utilized to determine how much detail is necessary. Critic can examine how much information is sufficient for the user and vary response by cutting off an unimportant, redundant goal during the course of planning. Furthermore, a focusing mechanism is introduced to ensure that the generated explanation is coherent. Focusing information is also used to decide on how to express each sentence, i.e., whether pronominalization can be applicable, which sentence form is more appropriate, an active voice or passive one, etc.
KeywordsUser Model Knowledge Structure Assembly Language Conceptual Frame Text Generation
Unable to display preview. Download preview PDF.
- Bresnan J (ed) (1982) The Mental Representation of Grammatical Relations. MIT Press, Cambridge, Mass.Google Scholar
- Cohen PR, Perrault CR (1979) Elements of a Plan-Based Theory of Speech Acts. Cognitive Science 3: 177–212Google Scholar
- Fillmore CJ (1968) The case for case. In: Bach E, Harms R (eds) Universals in Linguistic Theory. Holt Rinehart Winston, New York, 1–88Google Scholar
- McKeown KR (1985) Text Generation. Cambridge University Press, CambridgeGoogle Scholar
- Shapiro EY (1982) Algorithmic Program Debugging. MIT Press, Cambridge, Mass.Google Scholar
- Sidner CL (1983) Focusing in the Comprehension of Definite Anaphora. In: Brady M, Berwick RC (eds) Computational Models of Discourse. MIT Press, Cambridge, Mass., 267–330Google Scholar
- Simmons RF, Correira A (1979) Rule Forms for Verse, Sentences, and Story Trees. In: Findler NV (ed) Associative Networks: Representation and Use of Knowledge by Computers. Academic Press, New York, 363–392Google Scholar
- Uehara K, Kakiuchi T, Mikami O, Toyoda J (1986) Extended Prolog and Its Application to an Integrated Parser. In: Wada E (ed) Logic Programming '85. Springer, Berlin Heidelberg New York, 214–225Google Scholar
- Wilensky R, Arens Y, Chin D. (1984) Talking to UNIX in English: An Overview of UC. Comm. of the ACM 27: 574–593Google Scholar