The unbalanced motion caused by a peripheral vestibular dysfunction provokes spontaneous nystagmus. The major problem in Videonystagmography technique (VNG) is how to recognize a vestibular disorder. In this paper, a novel method is proposed for automatic extraction of Vestibulo-Ocular Response (VOR). The proposed methodology focuses on features reduction from VNG dataset in order to enhance the peripheral Vestibular Diseases (VD) diagnosis. The main contribution of the present work is the proposal of an automatic VD detection scheme which combines an improved feature characterization procedure and the Fuzzy C-means (FCM) classifier in order to study the latter’s ability in evaluating the vestibular dysfunction status within a reduced processing time. The use of VNG parameters gives interesting experimental results that show the effectiveness of the proposed method since up to 92% of classification accuracy was achieved, which is a significant rate as far as the experts' evaluation is concerned.