A Genetic Algorithm-based Multi-class Support Vector Machine for Mongolian Character Recognition

This paper proposes a hybrid genetic algorithm and support vector machine (GA-SVM) approach to address the Mongolian character recognition problem. As the character recognition problem can be considered as a multi-class classification problem, we devise a DAG-SVM classifier. DAG-SVM uses the One-Against-One technique to combine multiple binary SVM classifiers. The GA is used to select the multi-class SVM model parameters. Empirical results demonstrate that the GA-SVM approach is able to achieve good accuracy rate.