Restricting Answer Candidates Based on Taxonomic Relatedness of Integrated Lexical Knowledge Base in Question Answering

  • Received : 2016.08.12
  • Accepted : 2017.01.04
  • Published : 2017.04.01


This paper proposes an approach using taxonomic relatedness for answer-type recognition and type coercion in a question-answering system. We introduce a question analysis method for a lexical answer type (LAT) and semantic answer type (SAT) and describe the construction of a taxonomy linking them. We also analyze the effectiveness of type coercion based on the taxonomic relatedness of both ATs. Compared with the rule-based approach of IBM's Watson, our LAT detector, which combines rule-based and machine-learning approaches, achieves an 11.04% recall improvement without a sharp decline in precision. Our SAT classifier with a relatedness-based validation method achieves a precision of 73.55%. For type coercion using the taxonomic relatedness between both ATs and answer candidates, we construct an answer-type taxonomy that has a semantic relationship between the two ATs. In this paper, we introduce how to link heterogeneous lexical knowledge bases. We propose three strategies for type coercion based on the relatedness between the two ATs and answer candidates in this taxonomy. Finally, we demonstrate that this combination of individual type coercion creates a synergistic effect.



  1. J. Burger et al., "Issues, Tasks and Program Structures to Roadmap Research in Question & Answering," Document Understanding Conf. Roadmapping Documents, 2001, pp. 1-35.
  2. J. Chu-Carroll et al., "IBM's PIQUANT II in TREC 2004," Proc. TREC, Gaithersburg, MD, USA, Nov. 16-19, 2004, pp. 184-191.
  3. D. Moldovan et al., "The Structure and Performance of an Open-Domain Question Answering System," Proc. Annu. Meeting ACL, Hong Kong, China, Oct. 3-6, 2000, pp. 563-570.
  4. P.M. Ryu, M.G. Jang, and H.K. Kim, "Open Domain Question Answering Using Wikipedia-Based Knowledge Model," Inform. Process. Manage., vol. 50, no. 5, Sept. 2014, pp. 683-692.
  5. P.C. Chen, M.J. Zhuang, and C.J. Lin, "Using Wikipedia and Semantic Resources to Find Answer Types and Appropriate Answer Candidates Sets in Question Answering," Open Knowl. Base Question Answering Workshop COLING, Osaka, Japan, Dec. 2016.
  6. D.A. Ferrucci, "Introduction to 'This is Watson'," IBM J. Res. Develop., vol. 56, no. 3.4, May-June 2012, pp. 1:1-1:15.
  7. A. Lally et al., "Question Analysis: How Watson Reads a Clue," IBM J. Res. Develop., vol. 56, no. 3.4, 2012, pp. 2:1-2:14.
  8. J.W. Murdock et al., "Typing Candidate Answers Using Type Coercion," IBM J. Res. Develop., vol. 56, no. 3.4, May-June 2012, pp. 7:1-7:13.
  9. M.A. Pasca and S.M. Harabagiu, "High Performance Question/Answering," Proc. Annu. Int. ACM SIGIR, New Orleans, LA, USA, Sept. 2001, pp. 366-374.
  10. C.K. Lee et al., "Fine-Grained Named Entity Recognition Using Conditional Random Fields for Question Answering," Proc. Asia Conf. Inform. Retrieval Technol., Singapore, Oct. 16-18, 2006, pp. 581-587.
  11. S.J. Lim et al., "Domain-Adaptation Technique for Semantic Role Labeling with Structural Learning," ETRI J., vol. 36, no. 3, June 2014, pp. 429-438.
  12. C. Park et al., "Korean Coreference Resolution with Guided Mention Pair Model Using the Deep Learning," ETRI J., vol. 38, no. 6, Dec. 2016, pp. 1207-1217.
  13. T. Mikolov et al., "Distributed Representations of Words and Phrases and Their Compositionality," Proc. Int. Conf. Neural Inform. Process. Syst., Lake Tahoe, NV, USA, Dec. 5-10, 2013, pp. 3111-3119.
  14. D.W. Zhang et al., "Chinese Comments Sentiment Classification Based on Word2vec and $SVM^{perf}$," Expert Syst. Applicat., vol. 42, no. 4, Mar. 2015, pp. 1857-1863.
  15. A. Toral et al., "A Study on Linking Wikipedia Categories to Wordnet Synsets Using Text Similarity," Proc. Recent Adv. Natural Language Process., 2009, pp. 449-454.
  16. S. Fernando and M. Stevenson, "Mapping WordNet Synsets to Wikipedia Articles," LREC Conf, Turkey, May 2012, pp. 590-596.
  17. H.S. Choe, Construction and Application of Large-Scale Korean User-Word Intelligent Network, Ph.D. dissertation, University of Ulsan, Rep. of Korea, 2007.
  18. A.S. Yoon et al., "Construction of Korean WordNet 'KorLex 1.5'," J. KIISE: Softw. Applicat., vol. 36, no. 1, 2009, pp. 95-126.
  19. J. Chu-Carroll et al., "Textual Resource Acquisition and Engineering," IBM J. Res. Develop., vol. 56, no. 3.4, May-June 2012, pp. 4:1-4:11.
  20. G. Hirst and D. St-Onge, "Lexical Chains as Representations of Context for the Detection and Correction of Malapropisms," WordNet: An Electronic Lexical Database, Cambridge, MA, USA: MIT Press, 1998, pp. 305-332.
  21. C. Leacock and M. Chodorow, "Combining Local Context and WordNet Similarity for Word Sense Identification," WordNet: An Electronic Lexical Database, Cambridge, MA, USA: MIT Press, 1998, pp. 265-283.
  22. Z. Wu and M. Palmer, "Verbs Semantics and Lexical Selection," Proc. Annu. Meeting ACL, Las Cruces, New Mexico, June 27-30, 1994, pp. 133-138.
  23. P. Resnik, "Using Information Content to Evaluate Semantic Similarity," Proc. Int. Joint Conf. Artificial Intell., Montreal, Canada, Aug. 20-25, 1995, pp. 448-453.
  24. J.J. Jiang and D.W. Conrath, "Semantic Similarity Based on Corpus Statistics and Lexical Taxonomy," Proc. Conf. Res. Comput. Linguistics, Taiwan, 1997.
  25. D. Lin, "An Information-Theoretic Definition of Similarity," Proc. Int. Conf. Mach. Learning, July 24-27, 1998, pp. 296-304.
  26. C.K. Lee and M.G. Jang, "A Prior Model of Structural SVMs for Domain Adaptation," ETRI J., vol. 33, no. 5, 2011, pp. 712-719.

Cited by

  1. RNN based question answer generation and ranking for financial documents using financial NER vol.45, pp.1, 2017,