A novel maximal figure-of-merit (MFoM) learning approach to text categorization is proposed. Different from the conventional techniques, the proposed MFoM method attempts to integ...
A variational level set method is developed for the supervised classification problem. Nonlinear classifier decision boundaries are obtained by minimizing an energy functional tha...
Abstract. Multiple classifier systems based on neural networks can give improved generalisation performance as compared with single classifier systems. We examine collaboration in ...
Real world images often contain similar objects but with different rotations, noise, or other visual alterations. Vision systems should be able to recognize objects regardless of ...
Classifier fusion strategies have shown great potential to enhance the performance of pattern recognition systems. There is an agreement among researchers in classifier combination...
Amin Assareh, Mohammad Hassan Moradi, L. Gwenn Vol...