Comparing Performance of Interval Neutrosophic Sets and Neural Networks with Support Vector Machines for Binary Classification Problems

In this paper, the classification results obtained from several kinds of support vector machines (SVM) and neural networks (NN) are compared with our proposed classifier. Our approach is based on neural networks and interval neutrosophic sets which are used to classify the input patterns into one of the two binary class outputs. The comparison is based on several classical benchmark problems from UCI machine learning repository. We have found that the performance of our approaches are comparable to the existing classifiers. However, our approach has taken into account of the uncertainty in the classification process.