DOI QR코드

DOI QR Code

A three-step sentence searching method for implementing a chatting system

채팅 시스템 구현을 위한 3단계 문장 검색 방법

  • 전원표 (강원대학교 컴퓨터정보통신공학전공) ;
  • 송영길 (강원대학교 컴퓨터정보통신공학전공) ;
  • 김학수 (강원대학교 컴퓨터정보통신공학전공)
  • Received : 2012.12.27
  • Accepted : 2013.02.25
  • Published : 2013.03.31

Abstract

The previous chatting systems have generally used methods based on lexical agreement between users' input sentences and target sentences in a database. However, these methods often raise well-known lexical disagreement problems. To resolve some of lexical disagreement problems, we propose a three-step sentence searching method that is sequentially applied when the previous step is failed. The first step is to compare common keyword sequences between users' inputs and target sentences in the lexical level. The second step is to compare sentence types and semantic markers between users' input and target sentences in the semantic level. The last step is to match users's inputs against predefined lexico-syntactic patterns. In the experiments, the proposed method showed better response precision and user satisfaction rate than simple keyword matching methods.

기존 채팅 시스템은 일반적으로 사용자 입력 문장과 데이터베이스 내 목표 문장들 사이의 어휘 일치도에 기반을 둔 방법을 사용한다. 그러나 이러한 방법은 어휘 불일치 문제를 자주 일으킨다. 이러한 문제를 해결하기 위해 순차적으로 적용되는 3단계 문장 검색 방법을 제안한다. 첫 번째 단계는 어휘 수준에서 사용자 입력 문장과 목표 문장들 사이의 공통 키워드 열을 비교하는 것이다. 두 번째 단계는 의미 수준에서 사용자 입력 문장과 데이터베이스 내 문장들 사이의 문장 유형과 의미 표지를 비교하는 것이다. 마지막 단계는 미리 정의된 어휘-구문 패턴을 사용자 입력 문장과 매칭하는 것이다. 실험에서 제안된 방법은 단순 키워드 매칭 방법 보다 더 나은 응답 정확도와 사용자 만족도를 보였다.

Keywords

References

  1. J.-M. Seo and H.-K. Hwang, "A design and implementation of database for integrated data management on shipboard" Journal of the Korean Society of Marine Engineering, vol. 34, no. 34, pp. 1188-1194, 2010 (in Korean). https://doi.org/10.5916/jkosme.2010.34.8.1188
  2. Y.-S. Ha, "The prospects of intelligent robots for marine pollution response", Journal of the Korean Society of Marine Engineering, vol. 26, no. 4, pp. 408-414, 2002 (in Korean).
  3. Y.-S. Ha and S.-J. Oh, "Multi-sensor fusion as a sensing technique for system intelligence", Journal of the Korean Society of Marine Engineering, vol. 24, no. 2, pp. 151-157, 2000 (in Korean).
  4. J. Weizenbaum "ELIZA-A computer program for the study of natural language communication between man can machine", Communications of the Association for Computing Machinery, vol. 9, no. 1, pp. 36-46, 1996.
  5. J.-H. Kim, D.-S Jang, and H.-S Kim, "Statistical generation of korean chatting sentences using multiple feature information", Korean Journal of Cognitive Science, vol. 20, no. 4, pp. 421-437, 2009 (in Korean). https://doi.org/10.19066/cogsci.2009.20.4.002
  6. M. L. Mauldin, "ChatterBots, TinyMuds, and the turing test: entering the Loebner prize competition", Proceedings of the Twelfth National Conference on Artificial Intelligence, vol. 1, pp. 16-21. 1994.
  7. B. P. Schumaker, Ying Liu, Mark Ginsburg, and Hsinchun Chen, "Evaluating mass knowledge acquisition using the ALICE chatterbot: the AZ-ALICE dialog system", International Journal of Human-Computer Studies, vol. 64, no 11, pp. 1132-1140. 2006. https://doi.org/10.1016/j.ijhcs.2006.06.008
  8. Jason L. Hutchens, Michael D. Alder, "Introducing MegaHAL", NeMLaP3 / CoNLL98 Workshop on Human-Computer Conversation, Association for Computational Linguistics, pp. 271-274, 1998.
  9. D. Vrajitoru, "Evolutionary sentence building for chatterbots", Genetic and Evolutionary Computation Conference, pp. 315-321, 2003.
  10. D. Vrajitoru, J. Ratkiewicz, "Evolutionary sentence combination for chatterbots", The IASTED International Conference on Artificial Intelligence and Applications, pp. 287-292, 2004.
  11. J. R Curran, R. K. Wong, "Formalization of transformation-based learning", Australasian Computer Science Conference, pp. 51-57, 2000.