设为首页
加入收藏
A NEW METHOD FOR NON-LINEAR CLASSIFY AND NON-LINEAR REGRESSION Ⅰ :INTRODUCTION TO SUPPORT VECTOR MACHINE
-
摘要: 简要介绍了近年来倍受瞩目的一种处理高度非线性分类、回归等问题的计算机学习的新方法——支持向量机（SVM）方法；分析了这一方法的特点及其在数值预报产品释用及气象研究业务中的应用前景。SVM是一种有坚实理论基础的新颖的小样本学习方法。它基本上不涉及概率测度及大数定律等，因此不同于现有的统计方法。从本质上看，它避开了从归纳到演绎的传统过程，实现了高效的从训练样本到预报样本的“转导推理”(transductive inference)，大大简化了通常的分类和回归等问题。SVM的最终决策函数只由少数的支持向量所确定，计算的复杂性取决于支持向量的数目，而不是样本空间的维数，这在某种意义上避免了“维数灾”。
-
关键词:
- 机器学习;
- 支持向量机（SVM）;
- 模式识别;
- 回归估计;
- 天气预报
Abstract: A brief introduction to an increasingly popular machine learning technique, SVM (support vector machine) is presented for solving nonlinear classification and regression problems. Properties of SVM are discussed together with potentials of applying SVM to numerical weather forecast. SVM is a novel learning method that has solid theoretical basis and requires only small amount of sample. It does not rely on probability measures and Law of Large Numbers, hence is different from many other statistical methods. In essence, SVM smartly evades the traditional inference process from induction to deduction. Instead, it employs transductive inference from training sample to predicting sample, which greatly simplifies classification and regression problems. The decision function of SVM is only determined by a few support vectors. The complexity of computation relates to the number of support vectors rather than the dimension of the sample space. Thus, to some degree, SVM avoids the "curse of dimensionallty". -
[1] Vapnik V N.Statistical Learning Theory.John Wiley & Sons, Inc., New York, 1998. [2] Vapnik V N.The Nature of S tatisti cal Learning Theory.Springer Verlag, New York, 2000.(有中译本:张学工译.统计学习理论的本质.北京:清华大学出版社, 2000.) [3] Cristianini N and Shaw a-Taylor J.An Introduction of Support Vector Machines and Other Kernel based Learning M ethods.Camb ridge University Press, 2000. [4] Burges C J.A tutorial on support vector machines for pat t ern recognition.Data Mining an d Knowledge Discovery, 1998, 2 :127-167. [5] Courant R and Hilbert D, Method of Mathematical Physics, Volume I.Springer Verlag, 1953. [6] http://www.kernel-machines.org/ [7] Scholkopf B, Burges Ch-J C and Smola A J, edited.Advances in Kernel Methods-Support Vector Learning.M IT Press, Cambridge, 1999. -
下载:
图(6)
计量
- 摘要浏览量: 5258
- HTML全文浏览量: 680
- PDF下载量: 4238
- 被引次数: 0
