论文部分内容阅读
针对近红外(Near Infrared,NIR)光谱测量中的小样本问题。本文提出了一种集成最小二乘支持向量机(Ensemble Least Squares Support Vector Machine,ELS-SVM)新算法。首先使用随机子空间算法(Random Subspace Method,RSM)原始高维变量空间划分为若干个低维度的子空间,然后分别在各个子空间建立最小二乘支持向量机(LS-SVM)模型,最后构造一个集成结果来进行预测。针对一批柴油样本的实验结果表明,本法对柴油十六烷值的预测精度优于传统的LS-SVM方法。
For small sample problems in Near Infrared (NIR) spectroscopy. In this paper, a new algorithm of Ensemble Least Squares Support Vector Machine (ELS-SVM) is proposed. Firstly, the original high dimensional space is divided into several low dimensional subspaces using Random Subspace Method (RSM), and then LS-SVM models are established in each subspace. Finally, An integrated result to make predictions. Experimental results on a batch of diesel samples show that the prediction accuracy of diesel cetane by this method is superior to the traditional LS-SVM method.