Interest in the combination of statistical pattern recognition learning techniques and rule-based learning has grown in the last few years, and it appears as if it will be an important field in the future. In this paper, a novel system is presented that embeds the support vector machine (SVM) technique in a rule extraction framework. The proposed method extracts rules directly from the support vectors of a trained SVM using a modified sequential covering algorithm. The rules are generated based on an ordered search of the most discriminative features (as measured by interclass separation), which is achieved using SVM training. The performances of the rules achieved are evaluated by using the measured rates of true positives and false positives in an exhaustive classification test performed on a public dataset. The paper is well written, interesting, and gives a deep insight into the state of the art of rules extraction methods.