Support vector machine is a new learning method developed in recent years based on the foundations of statistical learning theory. It is gaining popularity due to many attractive features and promising empirical performance in the fields of nonlinear and high dimensional pattern recognition. TSVM (transductive support vector machine) takes into account a particular test set and tries to minimize misclassifications of just those particular examples. Compared with traditional inductive support vector machines, TSVM is often more practical and can give results with better performance. In this paper, a progressive transductive support vector machine is devised and the experimental results show that the algorithm is very promising on a mixed training set of a small number of unlabeled examples and a large number of labeled examples.