Abstract
We propose a new string kernel based on variable-length-don’t-care patterns (VLDC patterns). A VLDC pattern is an element of (Σ ∪ { ⋆ })*, where Σ is an alphabet and ⋆ is the variable-length-don’t-care symbol that matches any string in Σ*. The number of VLDC patterns matching a given string s of length n is O(22n). We present an O(n 5 ) algorithm for computing the kernel value. We also propose variations of the kernel which modify the relative weights of each pattern. We evaluate our kernels using a support vector machine to classify spam data.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Leslie, C.S., Eskin, E., Weston, J., Noble, W.S.: Mismatch string kernels for svm protein classification. In: Advance in Neural Information Processing Systems 15, pp. 1417–1424 (2002)
Cortes, C., Haffner, P., Mohri, M.: Rational kernels: Theory and algorithms. Journal of Machine Learning Research 5, 1035–1062 (2004)
Cortes, C., Kontorovich, L., Mohri, M.: Learning languages with rational kernels. In: Proceedings of the 20th Annual Conference on Learning Theory, pp. 349–364 (2007)
Lodhi, H., Saunders, C., Shawe-Taylor, J., Cristianini, N., Watkins, C.: Text classification using string kernels. Journal of Machine Learning Research 2, 419–444 (2002)
Inenaga, S., Bannai, H., Shinohara, A., Takeda, M., Arikawa, S.: Discovering best variable-length-don’t-care patterns. In: Lange, S., Satoh, K., Smith, C.H. (eds.) DS 2002. LNCS, vol. 2534, pp. 86–97. Springer, Heidelberg (2002)
Rahman, M.S., Iliopoulos, C.S., Lee, I., Mohamed, M., Smyth, W.F.: Finding patterns with variable length gaps or don’t cares. In: Chen, D.Z., Lee, D.T. (eds.) COCOON 2006. LNCS, vol. 4112, pp. 146–155. Springer, Heidelberg (2006)
Navarro, G., Raffinot, M.: Fast and simple character classes and bounded gaps pattern matching. Journal of Computational Biology 10(6), 903–923 (2003)
Boser, B.E., Guyon, I.M., Vapnik, V.N.: A training algorithm for optimal margin classifiers. In: COLT 1992: Proceedings of the fifth annual workshop on Computational learning theory, pp. 144–152. ACM, New York (1992)
Inenaga, S., Takeda, M., Shinohara, A., Hoshino, H., Arikawa, S.: The minimum DAWG for all suffixes of a string and its applications. In: Apostolico, A., Takeda, M. (eds.) CPM 2002. LNCS, vol. 2373, pp. 153–167. Springer, Heidelberg (2002)
Inenaga, S., Shinohara, A., Takeda, M., Bannai, H., Arikawa, S.: Space-economical construction of index structures for all suffixes of a string. In: Diks, K., Rytter, W. (eds.) MFCS 2002. LNCS, vol. 2420, pp. 341–352. Springer, Heidelberg (2002)
Narisawa, K., Bannai, H., Hatano, K., Takeda, M.: Unsupervised spam detection based on string alienness measures. In: Corruble, V., Takeda, M., Suzuki, E. (eds.) DS 2007. LNCS (LNAI), vol. 4755, pp. 159–172. Springer, Heidelberg (2007)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2008 Springer Berlin Heidelberg
About this paper
Cite this paper
Narisawa, K., Bannai, H., Hatano, K., Inenaga, S., Takeda, M. (2008). String Kernels Based on Variable-Length-Don’t-Care Patterns. In: Jean-Fran, JF., Berthold, M.R., Horváth, T. (eds) Discovery Science. DS 2008. Lecture Notes in Computer Science(), vol 5255. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-88411-8_29
Download citation
DOI: https://doi.org/10.1007/978-3-540-88411-8_29
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-88410-1
Online ISBN: 978-3-540-88411-8
eBook Packages: Computer ScienceComputer Science (R0)