The Research of Chinese Text Automatic Classification Based on Multiple

Abstract:

Support Vector Machines (SVMs) is a new technique for data mining. It has wide applications in various fields and is a research hot pot of the machine learning field, but, being applied to handling large-scale problems, SVMs needs longer t raining time and larger memory. It’s an effective way to solve large scale data processing in text classification with multiple classifier systems composed by multiple support vector machine classifiers. Based on the analysis of traditional parallel algorithms, this paper proposes an improved algorithm based on multiple SVMs. The experimental results indicate that the new algorithm works well in precision and recall rate in the condition that the speeds of classification increase remarkably. Compared with traditional algorithms, the classified accuracy is lower but is within the range for acceptance.

• Periodical: Applied Mechanics and Materials (Volumes 263-266)
• Pages: 1543-1548
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.263-266.1543
• Online since: December 2012
• Authors: Sheng Li Zhang
• Keywords: Multiple Classifier, Multiple SVMs, Parallel Algorithm, Text Classification

References

[1] Gu Yaxiang, Ding Shifei.:Advances of support vector machines(SVM). Computer Science, Vol.38(2)(2011),p:14~17.

[2] Qi Hengnian.:Support vector machines and application research overview.Computer Engineering, Vol.30(10)(2004),p:6~9.

[3] Wang Xiaodan,Wang Jiqin.:A Survey on support vector machines training and testing algorithms. Computer Engineering and Applications, Vol.13(10)(2004), p:75~78.

[4] Wang Guoshegn,Zhong Yixin.:Some new developments on support vector machin. Acta Electronica Sinica Vol.29(10)(2001), p:1396~1400.

[5] Ralaivola L, Flovenced.:Incremental support vector machine learning: a local approach proceedings of international conference on neural networks. Vienna, Austria, Vol.1(2001),p: 322~330

DOI: 10.1007/3-540-44668-0_46

[6] Li Yongli, Liu Yanheng, Xiao Jiantao.:Incremental Learning Algorithm Based on support vector machine. Journal of Jilin University(Science Edition), Vol.48(3)(2010),p:164~467.

[7] Yang Jing, Zhang Jianpei, Liu Daxin.:Research on incremental learning algorithm with multiple support vector machine classifiers. Journal of Harbin Engineering University, Vol.26(1)(2006),p:103~106.

[8] Tsing IW , Kwok JT ,Cheung PM.:Core vector machines :fastsvm training on very large data sets. Journal of MachineLearning Research , Vol.6(4)(2005),p:363~392.

[9] Yu Hong, Yang Jiang , Han Jiawei.:Making SVMs scalable to large data sets using hierarchical cluster indexing. Data Mining and Knowledge Discovery , Vol.11(3)(2005),p:295~321.

DOI: 10.1007/s10618-005-0005-7

[10] Collobert R,Bengio S,Bengio Y. A parallel mixture of SVMs for very large scale problem. Neural Computation ,Vol.14(7)(2002),p:1105~1114.

DOI: 10.1162/089976602753633402

[11] WenYimin,Wang Yaonan,Lu Baoliang.:Survey of applying support vector machines to handle large-scale problems. Computer Science, Vol.36(7)(2009),p:20~25.