Recent advances in Support Vector Machines

This special issue aims to present the related advances in Support Vector Machines and promote the research, development and applications of SVM by providing a high-level international forum for researchers and practitioners to exchange research results and share development experiences. Support Vector Machines (SVM), proposed by Vapink et al., is an effective machine learning method based on statistical learning theory. Compared to other machine learning methods, such as artificial neural networks, SVM can solve the problem of high dimension and local minima successfully, which makes it have better generalization ability. Due to the excellent performance of SVM, SVM has been successfully applied in many fields, such as pattern recognition, time series prediction, text classification and image processing. Submissions came from an open call for paper. A total of about 100 submissions were received, from which 22 papers were accepted for publication. Each submission went through an rigorous peer-review process. All the papers received at least two or three independent reviews, and have neither been submitted to nor published in any journals or conferences. The papers mainly cover the scope of SVM, but not limited to the scope. The papers are organized as follows: The paper “Using unsupervised clustering approach to train the support vector machine for text classification”, authored by Niusha Shafiabady, Lam Hong Lee, R. Rajkumar, Vish P Kallimani, Nik Ahmad Akram and Dino Isa, presents a new text classification approach. It uses the unsupervised approaches such as SelfOrganizing Maps (SOM) and Correlation Coefficient (CorrCoef) to group the unlabeled text documents and use the clustered text documents as the machine labeled training set for the Support Vector Machine (SVM). The paper “A graph-theoretic approach to 3D shape classification”, authored by Abdessamad Ben Hamza, introduce a 3D shape classification approach using graph regularized sparse coding in conjunction with the biharmonic distance map. The approach exploits both sparsity and dependence among the features of shape descriptors in a bid to design robust shape signatures that are effective in discriminating between shapes from different classes. In an effort to coherently capture the similarity between feature descriptors, it uses multiclass support vector machines for 3D shape classification on midlevel features that are learned via graph regularized sparse coding. The paper “Support vector machine based on hierarchical and dynamical granulation”, authored by Husheng Guo and Wenjian Wang, presents an improved granular support vector machine learning model based on hierarchical and dynamical granulation, namely, HD_GSVM. to solve the low learning efficiency and generalization performance problem of traditional granular support