Abstract
In this paper we report on new developments in our data-driven, non-linguistic, language-independent approach to Question Answering (QA). In particular, we describe a new implementation of the filter-model, which is used for answer typing, where we employ the Singular Value Decomposition (SVD) in a variation on the popular Latent Semantic Analysis technique. We also describe refinements to the open-source SVD code that we used which enable us to perform the SVD on arbitrarily large matrices. Finally, we discuss results from the TREC 2005 and TREC 2006 QA evaluations in which we applied these new techniques, and compare them to results achieved with our previous filter-model approach. In particular, we show that our new filter-model using the SVD achieves an average absolute gain of around 8% and an average relative gain of nearly 59% over our previous approach for top one answer accuracy. By using both approaches in combination we are able to increase the absolute gain to approximately 10% and the relative gain to 67%.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Whittaker, E., Furui, S., Klakow, D.: A Statistical Pattern Recognition Approach to Question Answering using Web Data. In: Proc. Cyberworlds (2005)
Whittaker, E., et al.: TREC2005 Question Answering Experiments at Tokyo Institute of Technology. In: Proc. 14th Text Retrieval Conference (2005)
Whittaker, E., Hamonic, J., Furui, S.: A Unified Approach to Japanese and English Question Answering. In: Proc. NTCIR-5 (2005)
Landauer, T., Laham, D., Foltz, P.: Learning human-like knowledge by singular value decomposition: a progress report, pp. 45–51. MIT Press, Cambridge (1998)
Cullum, J., Willoughby, R.: Lanczos Algorithms for Large Symmetric Eigenvalue Computations, ch. 5, vol. 1 Theory. Birkhauser, Boston (1985)
Whittaker, E., et al.: Monolingual Web-based Factoid Question Answering in Chinese, Swedish, English and Japanese. In: Proc. EACL (2006)
Bellegarda, J.: Latent Semantic Mapping. IEEE Magazine, 70–80 (2005)
Rhode, D.: SVDPACKC, http://tedlab.mit.edu/~dr/SVDLIBC/
Berry, M.: Large Scale Singular Value Decompositions. International Journal of Supercomputer Applications, 13–49 (1992)
Vector: Vector Software Library (1995-2003), http://www.vector.co.jp/
Google: Google Translate web translation, http://translate.google.com
TREC: TREC home, http://trec.nist.gov/
Academic Hallmarks: Knowledge Master Educational Software PO BOX 998 Durango, CO. 8130 (2002)
Hofmann, T.: Probabilistic Latent Semantic Analysis Uncertainty in Artificial Intelligence (1999)
Novak, J., et al.: NTCIR-6 CLQA Question Answering Experiment at the Tokyo Institute of Technology NTCIR-6 (2006)
Whittaker, E., et al.: CLEF2007 Question Answering Experiments at Tokyo Institute of Technology CLEF Working Notes (2007)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2008 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Whittaker, E., Novak, J., Heie, M., Imai, S., Furui, S. (2008). Using Singular Value Decomposition to Compute Answer Similarity in a Language Independent Approach to Question Answering. In: Tokunaga, T., Ortega, A. (eds) Large-Scale Knowledge Resources. Construction and Application. LKR 2008. Lecture Notes in Computer Science(), vol 4938. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-78159-2_24
Download citation
DOI: https://doi.org/10.1007/978-3-540-78159-2_24
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-78158-5
Online ISBN: 978-3-540-78159-2
eBook Packages: Computer ScienceComputer Science (R0)