Deep Non-Parallel Hyperplane Support Vector Machine for Classification

In the last few decades, deep learning based on neural networks has become popular for classification tasks, which combines feature extraction with tasks and always achieves satisfactory performance. Non-parallel hyperplane support vector machine (NPHSVM) aims at constructing two non-parallel hyperplanes to classify data extracted features are used be input NPHSVM. As NPHSVM, will greatly influence performance of model some extent. Therefore, in this paper, we propose a novel DNHSVM classification, generation seamlessly. Each is close its own class as far possible other classes, friendly samples easy classified. Experiments UCI datasets show effectiveness our proposed method, outperforms compared state-of-the-art algorithms.