Effective Data Classification via Combining Neural Networks and SVM
Han Liu, Xiaowu Xiao, LI Yan-xiang, Qiang Mi, Zhenwen Yang · 2019
In this paper, we propose an effective data classification method via combining neural networks and support vector machine (SVM). The neural networks are able to effectively extract data features. SVM is good at solving the small-sample classification problem. In order to combine the complementary advantages of neural networks and SVM, we first utilize neural works to extract data features, which are then classified with SVM. The proposed method is named as NN-SVM. We use the Iris dataset and the Lead isotope dataset to train and test NN-SVM. Experimental results show that NN-SVM outperforms neural networks based and SVM based classification algorithms.