Protein-RNA interactions play significant roles in a number of biological activities, such as protein synthesis, regulation of gene expression. Here we propose a hybrid RISP (RNA-interaction site prediction) method, using support vector machine (SVM) in conjunction with evolutionary information of amino acid sequences in terms of their position-specific scoring matrices (PSSMs) for prediction of RNA-binding sites. The results show that our RISP method has 72.2% net prediction (NP) (61.0% sensitivity and 83.3% specificity). When compared with previous studies, this novel method appears more accurate and better generalization abilities. RISP is freely available at http://grc.seu.edu.cn/RISP. Given a protein sequence, RISP decides whether residue in the protein is RNA-binding or not (optimal prediction), and gives the confidence value, 'high specificity' prediction and 'high sensitivity' prediction.