Advertisement

A travel consultation system: Towards a smooth conversation in Japanese

  • H. Suzuki
  • M. Kiyono
  • S. Kougo
  • M. Takahashi
  • S. Motoike
  • T. Niki
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 221)

Abstract

We have developed a travel consultation system which extracts user's requests through natural language (Japanese) conversation and answers destinations that fit the case.

The system has four parts, the parsing subsystem, the user model subsystem, the sentence generator, and the database retriever. All subsystems are written in PROLOG.

The distinctive characters of conversations which our system makes are:
  • smoothness and kindness,

  • that the system can be a topic introducer, and

  • efficiency.

How the system works is as follows:

First, the system parses Japanese sentences that the user inputs. The input sentences are interpreted as user's requests, not as the answer for the previous question. So he can add any requests other than the questioned one. Next, the system guesses the user's type, evaluating user models that the system has. Each user model represents the expected type of users. Then it generates a next question which that type of users would care. This parse-guess-generate sequence are continued till the topics which the user cares are exhausted. At last the system searches the database and shows places that satisfy all requests of the user.

Keywords

User Model Transformation Rule Horn Clause Input Tree Sentence Generator 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Barwise J, Perry J (1983) Situations and Attitudes. MIT Press, Cambridge, Mass.Google Scholar
  2. Bobrow DG, Kaplan RM, Kay M, Norman DA, Thompson H, Winograd T (1977) GUS, A Frame-Driven Dialog System. Artificial Intelligence 8: 155–173CrossRefGoogle Scholar
  3. Carbonell JR (1970) AI in CAI: An Artificial Intelligence Approach to Computer-Assisted Instruction. IEEE transaction on Man-Machine Systems MMS-11: 190–202Google Scholar
  4. Coulthard M (1977) An Introduction to Discourse Analysis. Logman, London Grice HP (1975) Logic and Conversation. In: Morgan JL (ed) Syntax and Semantics III: Speech Acts. Academic Press, New York, p 41Google Scholar
  5. Inoue K (1976) Henkei Bunpou to Nihongo ('Transformational Grammar and Japanese'). Taishukan, Tokyo (in Japanese)Google Scholar
  6. Kaplan RM, Bresnan J (1982) Lexical-Functional Grammar. In: Bresnan J (ed) Mental Representation of Grammatical Relations. MIT Press, Cambridge, Mass., p 173Google Scholar
  7. Mizutani S, Ishiwata T, Ogino T, Kaku N, Kusanagi Y (1983) Bunpou to Imi I ('Grammar and Semantics I'). Asakura, Tokyo (in Japanese)Google Scholar
  8. Neustupný JV (1982) Gaikokujin-tono Communication ('Communication with Foreigners'). Iwanami, Tokyo (in Japanese)Google Scholar
  9. Shibatani M (1978) Nihongo-no Bunseki ('Analysis of Japanese'). Taishukan, Tokyo (in Japanese)Google Scholar
  10. Suzuki H (1984) Nihongobun-no Imi-no Joukyouimirontekina Kijutsu ('Event Types Represent Meanings of Japanese Sentences'). Information Processing Society of Japan, WGNL 42-3 (in Japanese)Google Scholar
  11. Suzuki H, Kiyono M, Kougo S (1985) User-no Kaiwa-no Kata-wo Mochii-ta Shitsumon Outou System ('selecting Topics in a Dialog according to Users Type'). Information Processing Society of Japan, WGNL 49-3 (in Japanese)Google Scholar
  12. Takahashi M, Suzuki H, Kiyono M (1985) Production System wo Mochii-ta Nihongo Bunseisei ('Generation of Japanese Sentences by Production'), Proceedings of the 30th semiannual meeting of Information Processing Society of Japan: 1675–1676 (in Japanese)Google Scholar
  13. Teramura H (1982) Nihongo-no Syntax to Imi I ('syntax and Semantics of Japanese I'). Kuroshio, Tokyo (in Japanese)Google Scholar
  14. Yasukawa H (1984) LFG System in Prolog. Proceedings of the 10th International Conference on Computational Linguistics: 358–361Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1986

Authors and Affiliations

  • H. Suzuki
    • 1
  • M. Kiyono
    • 1
  • S. Kougo
    • 1
  • M. Takahashi
    • 1
  • S. Motoike
    • 1
  • T. Niki
    • 1
  1. 1.Tokyo Systems Research DepartmentMatsushita Electric Industrial Co., Ltd.TokyoJapan

Personalised recommendations