期刊文献+

基于类边界壳向量的快速SVM增量学习算法 被引量:8

Fast SVM incremental learning algorithm based on between-class convex hull vectors
下载PDF
导出
摘要 为进一步提高SVM增量训练的速度,在有效保留含有重要分类信息的历史样本的基础上,对当前增量训练样本集进行了约简,提出了一种基于类边界壳向量的快速SVM增量学习算法,定义了类边界壳向量。算法中增量训练样本集由壳向量集和新增样本集构成,在每一次增量训练过程中,首先从几何角度出发求出当前训练样本集的壳向量,然后利用中心距离比值法选择出类边界壳向量后进行增量SVM训练。分别使用人工数据集和UCI标准数据库中的数据进行了实验,结果表明了方法的有效性。 To reduce the computational cost of the incremental learning,a fast Support Vector Machine(SVM) incremental learning algorithm is proposed,and the definition of the between-class convex hull vector is given.The given algorithm is based on utilizing the result of the previous training effectively and retaining the most important samples for the incremental learning to reduce the computational cost.In the process of incremental learning,the convex hull vectors of the previous training and the newly added samples constitute the current training sample set,the current training sample set is pre-extracted from the geometric point of view by using the convex hulls algorithm,the central distance ratio method is used to obtain the between-class convex hull vectors before the SVM incremental training.Experiments prove that the given algorithm has better classification performance.
出处 《计算机工程与应用》 CSCD 北大核心 2010年第23期185-187,248,共4页 Computer Engineering and Applications
基金 国家自然科学基金No.50505051 陕西省自然科学研究计划No.2007F19~~
关键词 支持向量机 增量学习 壳向量 Support Vector Machine(SVM) incremental learning convex hull vector
  • 相关文献

参考文献10

  • 1Vapnik V N.The nature of statistical learning theory[M].2nd ed. New York: Springer-Verlag, 2000.
  • 2Zhang Yizhuo.Constructing multiple support vector machines ensemble based on fuzzy integral and rough reducts[C]//Proceedings of 2nd IEEE Conference on Industrial Electronics and Applications, 2007:1256-1259.
  • 3Kivinen J, Smola A J,Williamson R C.Online learning with kernels[C]//Proc of Advances in Neural Information Processing Systems, Cambridge, MA, 2002.
  • 4Cauwenberghs G, Poggio T.Incremental and decrementa! support vector machine leaming[J].Machine Learning ,2001,44( 13 ) :4098-4151.
  • 5Syed N, Liu H, Sung K K.Incremental learning with support vector machines[C]//Proc of Workshop on Support Vector Machines at the International Joint Conference on Artificial Intellgence(IJCAI-99), Stockholm, Sweden, 1999.
  • 6萧嵘,王继成,孙正兴,张福炎.一种SVM增量学习算法[J].南京大学学报(自然科学版),2002,38(2):152-157. 被引量:24
  • 7李东晖,杜树新,吴铁军.基于壳向量的线性支持向量机快速增量学习算法[J].浙江大学学报(工学版),2006,40(2):202-206. 被引量:16
  • 8Barber C B, Dobkin D P, Huhdanpaa H T.The quickhull algorithm for convex hulls[EB/OL].http ://www.qhull.org.
  • 9Bennett K P,Bredensteiner E J.Duality and geometry in SVM classifiers[M].San Francisco,CA:Morgan Kaufmann,2000.
  • 10Keerthi S S, Shevade S K, Bhattacharyya C, et al.A fast iterative nearest point algorithm for support vector machine classifier design[J].IEEE Transaction on Neural Network, 2000, 11 (1): 124-136.

二级参考文献15

  • 1XIAO Rong,WANG Ji-cheng,ZHANG Fa-yan.An approach to incremental SVM learning algorithm[C]//ICTAI2000.Vancouver,British Columbia,Canada:IEEE,2000:268-273.
  • 2SYED N,LIU H,SUNG K K.Incremental learning with support vector machines[C]//IJCAI 1999.Stockholm:Morgan kaufmann publishers,1999:352-356.
  • 3]KRISTIN P B,ERIN J B.Geometry in learning[R].New York:Rennselaer Polytechnic Institute,1996.
  • 4.[EB/OL].http://www.bupa.co.uk/,.
  • 5Vapnik V N 张学工.统计学习理论的本质[M].北京:清华大学出版社,2000..
  • 6Vapnik V. The Nature of statistical learning theory. New York: Springer-Verlag,1995. 5-13.
  • 7Christopher J C B. A tutorial on support vector machines for pattern recognition. Knowledge Discovery Data Mining, 1998, 2(2): 235-244.
  • 8Vapnik V, Levin E, Cun Y Le. Measuring the VC-dimension of a learning machine. Neural Computation, 1994, 6(5): 851-876.
  • 9Vapnik V. Statistical learning theory. New York:Wiley, 1998. 21-22.
  • 10Vapnik V, Golowich S, Smola A. Support vector method for function approximation, regression estimation, and signal processing. Mozer M, Jordan M, Petsche T. Advances in Neural Information Processing Systems 9. Cambridge: MIT Press, 1997: 281-287.

共引文献35

同被引文献61

引证文献8

二级引证文献34

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部