SVM与ANN的比较.pdfVIP

基于 ANN 与 SVM 的分类和回归比较研究 1,2 1 林关成 , 李亚安 （1.西北工业大学 航海学院,陕西 西安 710072;2.渭南师范学院 传媒工程系,陕西 渭南 714000） 摘 要：支持向量机是一种基于统计学习理论的新型机器学习方法，广泛应用于解决分类和回归问题。由于其 出色的学习性能，已成为当前国际机器学习界的研究热点。介绍了支持向量机与人工神经网络的基本原理，从 理论上对人工神经网络与支持向量机的学习方法进行了分析与比较，并对二者进行了分类和回归方面的性能仿 真。结果表明，虽然支持向量机和人工神经网络模型都表现了较好的分类能力，但支持向量机的回归能力高于 人工神经网络方法，且在计算机速度上有人工神经网络无法比拟的优势。 关键词：人工神经网络；支持向量机；统计学习理论；分类；回归 中图分类号：TP18 文献标识码：A 文章编号： Research on the Comparison Artificial Neural Network with Support Vector Machine for Classification and Regression 1,2 1 Lin Guan-cheng , Li Ya-an ( 1.College of Marine, Northwestern Polytechnical University, Xi an,710072, China; 2.Department of Media and Communications Engineering, Weinan Teachers College, Weinan 714000，Shaanxi ) Abstract: Support vector machine (SVM) is a newly developed machine learning algorithm. Based on statistic learning theory, it has been successfully used for solving classification and regression problems. Because of their excellent learning performance, it has become the hotspot of machine learning. The paper firstly deals with the introduction of the basic theory of SVM and artificial neural network (ANN) in brief. And then it analyses and compares the learning ability of SVM and ANN. Simulation results indicate that both methods are able to do classification and regression with robustness. Furthermore the SVM is better than ANN in the ability of regressing, especially in computing speed. Keywords: ANN ；SVM；Statistic learning theory；Classification ；Regression 1 引言 基于数据的机器学习是现代智能技术中的