Abstract
In Castro et al. (Mathematics without boundaries: surveys in pure mathematics. Springer, New York, 2014), authors proposed a method, called Aveiro discretization method, in their paper to determine whether a given function belongs to a certain reproducing kernel Hilbert space or not, depending on finite data of the given function. The method is effective for many cases where the data are noise free. However, it will loose efficacy if the data are noise corrupted. To deal with noise-corrupted data case, we combine two support vector regression (SVR) algorithms and Averio discretization method, respectively, to determine a given function whether belongs to a certain reproducing kernel Hilbert space. In the text, our approach is phrased as the SVR-based Aveiro discretization method which includes two algorithms, the SVR-Aveiro discretization algorithm and LS-SVR-Aveiro discretization algorithm. The proposed approach is compared with Aveiro discretization method and Linear-SVR-Aveiro discretization algorithm which is combined of linear-SVR and Aveiro discretization method; our approach shows promising results.
Similar content being viewed by others
References
Castro LP, Fujiwara H, Rodrigues MM, Saitoh S (2012) A new discretization method by means of reproducing kernels. In: Hung Son L, Tutscheke W (eds) Interactions between real and complex analysis. Sci. and Tech. Publication House, Amsterdam, pp 185–223
Castro LP, Fujiwara H, Rodrigues MM, Saitoh S, Tuan VK (2014) Aveiro discretization method in mathematics: a new discretization principle. In: Pardalos P, Rassias TM (eds) Mathematics without boundaries: surveys in pure mathematics. Springer, New York
Castro LP, Fujiwara H, Qian T, Saitoh S (2014) How to cath smoothing properties and analyticity of functions by computers? In: Pardalos P, Rassias TM (eds) Mathematics without boundaries: surveys in pure mathematics. Springer, New York
Hsu CW, Chang CC, Lin CJ (2003) A practical guide to support vector classification. Technical report, Department of Computer Science and Information Engineering, National Taiwan University, Taipei. http://www.csie.ntu.edu.tw/cjlin/libsvm/
Keerthi SS, Lin C-J (2003) Asymptotic behaviors of support vector machines with Gaussian kernel. Neural Comput 15(7):1667–1689
Mattera D, Haykin S (1999) Support vector machines for dynamic reconstruction of chaotic systems. In:Sch\(\ddot{o}\)kopf S, Smola B (eds) Advances in kernel methods. MIT Press, New York
Rojo-álvarez JL, Camps-Valls G, Martinez-Ramón M, Soria-Olivas E, Figueiras-Vidal AR (2005) Support vector machines framework for linear signal processing. Signal Process 85(12):2316–2326
Saitoh S (1997) Integral transforms, reproducing kernels and their applications. In: Pitman res. notes in math. series, vol 369. Addison Wesley Longman, Harlow
Saitoh S (2010) Theory of reproducing kernels: applications to approximate solutions of bounded linear operator functions on Hilbert spaces. Amer. Math. Soc. Transl. Ser. 2, vol 230. Amer. Math. Soc., Providence
Suykens JAK, Vandewalle J (1999) Least squares support vector machine classifiers. Neural Process Lett 9(3):293–300. Lirias number: 218716
Vapnik V (1995) The nature of statistical learning theory. Springer, New York
Ye J, Xiong T (2007) SVM versus least squares SVM, AISTATS. In: JMLR proceedings, vol 2. JMLR.org, pp 644–651
Zhang P, Peng J (2004) SVM vs regularized least squares classification. In: ICPR, pp 176–179
Acknowledgments
The author would like to sincerely thank professor Tao Qian for pointing out this topic and discussing with the author. The author would like to thank sincerely the reviewers for their comments, which helped improve the paper significantly.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by V. Loia.
Rights and permissions
About this article
Cite this article
Mo, Y. Applications of SVR to the Aveiro discretization method. Soft Comput 19, 1939–1951 (2015). https://doi.org/10.1007/s00500-014-1379-5
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-014-1379-5