Direct acyclic graph based multi-class twin support vector machine for pattern classification
Divya Tomar, Sonali Agarwal · 2015
In this paper, we propose a novel Multi-class Twin Support Vector Machine (MTWSVM) classifier on the basis of Direct Acyclic Graph (DAG) approach. MTWSVM is the multi-class extension of the recently proposed binary Twin Support Vector Machine (TWSVM) classifier. The optimization problems of the proposed classifier are solved by the Successive Over Relaxation (SOR) technique which speed up the training phase. The performance of the proposed classifier is compared with the existing approaches and validated against ten benchmark datasets. Further, we have investigated the efficiency of proposed classifier for Handwritten Digits recognition application. The effectiveness of the proposed classifier over existing approaches is demonstrated with the help of experimental results.