Fuzzy-rough neural networks for vowel classification

While designing radial basis function neural networks for classification, fuzzy clustering is often used to position the hidden nodes in the input space. The main assumption of the clustering is that similar inputs produce similar outputs. In other words, it means that any two input patterns from the same cluster must be from the same class. Generalization is possible in the radial basis function neural networks due to this similarity property. In many real life applications, however, two patterns from the same cluster belong to different classes, and hence, classification based on mere similarity property is inadequate. This problem arises because the available features are not sufficient to discriminate the classes. It implies that the fuzzy clusters generated by the input features have rough uncertainty. This paper proposes a fuzzy-rough set based network which exploits fuzzy-rough membership functions to reduce this problem. The proposed network is theoretically a powerful classifier as it is equivalent to a universal approximator. Moreover, its activity is transparent as it can easily be mapped to a Takagi-Sugeno type fuzzy rule base system. The efficacy of the proposed method is studied on a vowel recognition problem.