论文部分内容阅读
支持向量机(SVM)是一种基于统计学习理论,针对小样本学习问题的通用学习算法,它采用结构风险最小化(Structural risk minimization,SRM)准则,大大提高了模型的泛化能力,成功地解决了神经网络的过学习问题。目前主要应用在模式识别领域,在工业过程中的应用相对较少。本文首先从理论研究、算法结构、参数选择和扩展SVM 4个方面详细介绍了近些年来支持向量机的研究进展;然后对SVM在工业过程中的应用现状进行分析,指出进一步研究的方向。
Support Vector Machine (SVM) is a general learning algorithm based on statistical learning theory and for small sample learning. It adopts the Structural Risk Reduction (SRM) criterion and greatly improves the generalization ability of the model. Successfully Solve the neural network over-learning problems. Currently mainly used in the field of pattern recognition, the application of industrial processes is relatively small. In this paper, the research progress of support vector machines in recent years is introduced in detail from the aspects of theoretical research, algorithm structure, parameter selection and extended SVM. Then, the application status of SVM in industrial process is analyzed and the direction of further research is pointed out.