RNA 5-methylcytosine (m5C) plays an important role in numerous biological processes. Accurate identification of the m5C site is helpful for a better understanding of its biological functions. However, the drawbacks of the experimental methods available preclude progress towards the identification of the m5C site. As an excellent complement to experimental techniques, computational methods will facilitate the identification of m5C. In the present study, a support vector machine based-method is proposed to identify m5C sites in Homo sapiens. In this method, RNA sequences are encoded using the pseudo dinucleotide composition in which three RNA physiochemical properties are incorporated. It was observed by the jackknife cross-validation that the overall success rate achieved by the proposed model is 90.42%. This result indicates that the proposed model holds the potential to become a useful tool for the identification of m5C sites.