Approximating support vector machine with artificial neural network for fast prediction

作者：

Highlights：

• Hybrid neural network (HNN), a method to accelerate prediction speed of support vector machine (SVM) is proposed.

• The proposed method approximates SVM using artificial neural network (ANN).

• The proposed method yields much faster prediction speed without compromising prediction accuracy.

• The application of this method can improve practical usability of SVM.

摘要

•Hybrid neural network (HNN), a method to accelerate prediction speed of support vector machine (SVM) is proposed.•The proposed method approximates SVM using artificial neural network (ANN).•The proposed method yields much faster prediction speed without compromising prediction accuracy.•The application of this method can improve practical usability of SVM.

论文关键词：Support vector machine,Artificial neural network,Hybrid neural network,Approximation,Run-time speed

论文评审过程：Available online 22 February 2014.

论文官网地址：https://doi.org/10.1016/j.eswa.2014.02.025