Incremental Regressive Learning Algorithm of Support Vector Machine and its Application

Shi Hua Zhang; Xi Long Qu; Xue Ye Wang

doi:10.4028/www.scientific.net/AMR.216.301

Paper Titles

Band Gaps of Two-Dimensional Square-Lattice Photonic Crystal with Button-Shaped Dielectric Rods
p.285

Design of a Weight Detection System for NI ELVIS II Based on LabVIEW
p.290

An Improved Image Encryption Algorithm Based on a New Two-Dimensional Map
p.293

A Security Method for Multi-Sensor Fused Image
p.297

Incremental Regressive Learning Algorithm of Support Vector Machine and its Application
p.301

Effects of Argon-Oxygen Ratio and Sputtering Power on Optical Properties of ZnO:Al Thin Films Fabricated by RF Magnetron Sputtering
p.307

Crystal Growth, Optical Spectra and Thermal Properties of Na₂Ni(SO₄)₂·4H₂O Crystal
p.312

A New Modeling on Information Transmission of Weak Electrical Signals in the Nutrition Growth of Plant
p.316

Electrochemical Transient Behaviors of Tubular Steel in Buffered Acidic Solution Dissolved with Hydrogen Sulfide and Carbon Dioxide
p.321

HomeAdvanced Materials ResearchAdvanced Materials Research Vol. 216Incremental Regressive Learning Algorithm of...

Incremental Regressive Learning Algorithm of Support Vector Machine and its Application

Abstract:

There is no incremental learning ability for the traditional support vector machine (SVM) and there are all kind of merits and flaws for usually used incremental learning method. Normal SVM is unable to train in large-scale samples, while the computer’s memory is limited. In order to resolve this problem and improve training speed of the SVM, we analyze essential characteristic of SVM and bring up the incremental learning algorithm of SVM based on regression of SVM related to SV (support vectors). The algorithm increases the speed of training and can be able to learning with large-scale samples while its regressive precision loses fewer. The experiments show that SVM performs effectively and practically. Its application to prediction of the transition temperature (T_g) for high molecular polymers show that this model (R²=0.9427) proved to be considerably more accurate compared to a ANNs regression model (R²=0.9269).

You might also be interested in these eBooks

Optical, Electronic Materials and Applications

View Preview

Info:

Periodical:

Advanced Materials Research (Volume 216)

Pages:

301-306

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.216.301

Citation:

Cite this paper

Online since:

March 2011

Authors:

Shi Hua Zhang, Xi Long Qu, Xue Ye Wang

Keywords:

ANNs Regression, Estimation of Regression, Incremental Learning, Support Vector Machine (SVM)

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] V. Vapnik. Statistical Learning Theory. Wiley, Chichester, GB, (1998).

Google Scholar

[2] H. Drucher, C.J. Burges, L. Kaufman, et al. Support vector regression machines, The MIT Press, MA, (1997).

Google Scholar

[3] Chapelle O, Vapnik V, Bacsquest O, et al. Choosing multiple parameters for support vector machines [J]. Machine Learning, 2002, 46(1): 131-159.

Google Scholar

[4] Elizondo D. The linear separability problem: Some testing methods, IEEE Transactions on Neural Networks, 2006(02).

DOI: 10.1109/tnn.2005.860871

Google Scholar

[5] A. Afantitis and G. Melagraki, Quantitative Structure-Property Relationship (QSPR) Correlation of Glass Transition Temperatures of High Molecular Weight Polymers, J. Molecular Structure 716(2005), p.193.

DOI: 10.1016/j.theochem.2004.11.021

Google Scholar

[6] Katritzky, A. R.; Sild, S.; Lobanov, V.; Karelson, M. Quantitative Structure-Property Relationship (QSPR) Correlation of Glass Transition Temperatures of High Molecular Weight Polymers. J. Chem. Inf. Comput. Sci. 1998, 38, 300-304.

DOI: 10.1021/ci9700687

Google Scholar