QA System Metis Based on Web Searching and Semantic Graph Matching

  • Dongli Han
  • Yuhei Kato
  • Kazuaki Takehara
  • Tetsuya Yamamoto
  • Kazunori Sugimura
  • Minoru Harada
Part of the IFIP International Federation for Information Processing book series (IFIPAICT, volume 228)

Abstract

We have developed a question-answering system Metis with natural-language interface. Metis generates the answer to a question by comparing the semantic graph of the question sentence with sentences discovered on the Internet as knowledge source. Specifically, we first get a set of semantic frames for the question sentence, as the output from a semantic analysis system, SAGE, Then we extract several keywords from all semantic frames using SVM. After that we search the Web to find knowledge sentences based on the keywords and input each knowledge sentence into SAGE in order to get its semantic graphs similarly. Finally, the similarities between the semantic graph of the question sentence and that of each knowledge sentence are calculated to determine the most reliable knowledge sentence, in which a constituent is chosen as the answer to the question. An experiment to examine the effectiveness of our method showed that 65% of the questions for which suitable knowledge sentences had been found were replied correctly.

Key words

Natural Language Processing Question Answering System Semantic Analysis EDR-Dictionary 

References

  1. 1.
    Endo, T., and Fukumoto, J.: QA System using Classified Type of Named Entity. IPSJ SIG Notes. NL-159, pp.25–30 (2004).Google Scholar
  2. 2.
    Kurata, G., Okazaki, Naoki., and Ishizuka, Mitsuru.: Question Answering System with Graph Structure from Dependency Analysis. IPSJ SIG Notes. NL-158, pp.69–74 (2003).Google Scholar
  3. 3.
    Sasaki, Y., Isozaki, H., Suzuki, J., Kokuryou, K., Hirao, T., Kazawa, H., and Maeda, E.: SAIQA-II: A Trainable Japanese QA System with SVM (Natural Language Processing), Transactions of-Information Processing Society of Japan. Vol. 45, No. 2, pp.635–646 (2004).Google Scholar
  4. 4.
    Murata, M., Utiyama, M., and Isahara, H.: Question Answering System Using Similarity-Guided Reasoning, IPSJ SIG Notes. NL-135, pp 181–188 (2000).Google Scholar
  5. 5.
    Maezawa, T., Menrai, M., Ueno, M., Han D., and Harada., M.: Improvement of the Precision of the Semantic Analysis System SAGE, and Generation of Conceptual Graph, Proceedings of the 66th National Convention of IPSJ, 2-6U-05, pp. 177–178(2004).Google Scholar
  6. 6.
    Harada, M., Tabuchi, K., Oono, H.: Improvement of Speed and Accuracy of Japanese Semantic Analysis System SAGE and Its Accuracy Evaluation by Comparison with EDR Corpus. Transactions of Information Processing Society of Japan. Vol.43, No. 9, pp.2894–2902 (2002).Google Scholar
  7. 7.
    Harada, M, MIZUNO, T.: Japanese Semantic Analysis System SAGE using EDR. Transactions of the Japanese Society for Artificial Intelligence. Vol.16, No.l, pp.85–93 (2001).CrossRefGoogle Scholar
  8. 8.
    Sowa, J.: Conceptual Structures, Information Processing in Mind and Machine, Addison-Wesley, Reading, MA (1984).MATHGoogle Scholar
  9. 9.
    Quiz Millionaire, Fuji Television. (2002).Google Scholar

Copyright information

© International Federation for Information Processing 2006

Authors and Affiliations

  • Dongli Han
    • 1
  • Yuhei Kato
    • 2
  • Kazuaki Takehara
    • 2
  • Tetsuya Yamamoto
    • 2
  • Kazunori Sugimura
    • 2
  • Minoru Harada
    • 2
  1. 1.Department of Computer Science and System AnalysisNINON UniversityJapan
  2. 2.Department of Integrated Information TechnologyAoyama Gakuin UniversityJapan

Personalised recommendations