Optimal Boundary SVM Incremental Learning Algorithm

Abstract:

Support vectors (SVs) cant be selected completely in support vector machine (SVM) incremental, resulting incremental learning process cant be sustained. In order to solve this problem, the article proposes optimal boundary SVM incremental learning algorithm. Based on in-depth analysis of the trend of the classification surface and make use of the KKT conditions, selecting the border of the vectors include the support vectors to participate SVM incremental learning. The experiment shows that the algorithm can be completely covered the support vectors and have the identical result with the classic support vector machine, it also saves lots of time. Therefore it can provide the conditions for future large sample classification and incremental learning sustainability.

• Periodical: Applied Mechanics and Materials (Volumes 347-350)
• Pages: 2957-2962
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.347-350.2957
• Online since: August 2013
• Authors: Jian Cao, Shi Yu Sun, Xiu Sheng Duan
• Keywords: Increamental Learning, KKT Condition, Support Vector Machine (SVM)

References

[1] Dongying Bai, Xiaodan Wang, Hong-da Zhang, Wen Quan. Concept drift on SVM incremental learning and its disposals[J]. Computer Engineering and Design, 2008, 5(4): 2619-2621.

[2] Ying Zhao. Incremental Learning Algorithm of Support Vector Machine Based on Vector Projection[D]. Harbin: Harbin engineering university, (2007).

[3] Yinggang Zhao, Qinming He. An Incremental Learning Algorithm Based on SuPPort Vector Domain Classifier[C]. Proe. 5th IEEE International Conference on Cognitive Informatics [ICCI'06], 2006: 805-809.

DOI: 10.1109/coginf.2006.365593

[4] Ye Liu, Zebing Wang, Yan Feng, Hongying Gu. DoS Intrusion Detection Based on Incremental Learning with Support Vector Machines[J]. Computer Engineering, 2006, 32(4): 179-180.

[5] Yueyang Teng, Huanwen Tang, Haixia Zhang. A New Algorithm to Incremental Learning with Support Vector Machine[J]. Computer Engineering and Applications, 2004, 36(5): 77-80.

[6] SyedN, LiuH, SungK. Incremental learning with support vector machines[C]. Proceedings of the Workshop on Support Vector Machines at the International Joint Conference on Artificial Intelligence(IJCAI299). Stockholm, Sweden: Morgan Kaufmann, 1999. 876-892.

[7] Weida Zhou, Li Zhang, Licheng Jiao. An Analysis of SVMs Generalization Performance[J]. Acta Electronica Sinica, 2001，29(5): 590-594.

DOI: 10.1109/icosp.2002.1179989

[8] Wenhua Zeng, Jian Ma. A Novel Approach to Incremental SVM Learning Algorithm[J]. Journal of Xiamen University(Natural Science), 2009, 41(6) : 687-691.

[9] Katagiri S, Abe S. Incremental training of support vector machines using hyperspheres[J]. Pattern Recognition Letters, 2006, 27(13): 1495-1507.

DOI: 10.1016/j.patrec.2006.02.016

[10] Chongming Wu, Xiaodan Wang, Dongying Bai, ZHANG Hong-da. Fast SVM incremental learning algorithm based on between-class convex hull vectors[J]. Computer Engineering and Applications, 2010, 46(23): 185-187, 248.

[11] Fa Zhu, Ning Ye, Dongyin Pan, Wen Ding. Incremental support vector machine approach based on minimum sumple plane distance[J]. Computer Engineering and Design, 2012, 33 (1): 346-350.