Support vector machines is a blackbox model whose knowledge is concealed in the decision function. This has not only weakened the confidence of users in building intelligent systems using support vector machines techniques, but also hindered the application of support vector machines to data mining. Since extracting rules from support vector machines help to solve those problems, this area is becoming a hot topic in both machine learning and intelligent computing communities. In this paper, the typical algorithms for rule extraction from support vector machines are introduced, and some issues valuable for future exploration in this area are indicated.