Abstract
With the beginning of revolution in education E-learning become increasingly popular. It is widespread among people while E-learning platform is used by more people to publish their own resources freely. As a result, a large number of document data need to be processed. Text classification is a key technology to this problem. In this paper, we propose a frame based on semi-supervised NMF and SVM for text classification. The method of semi-supervised NMF can work out a so called open set problem well. And we also proposed an accelerated version of the algorithm base on CUDA and GPU parallel architecture. These methods can work out the problems well. And finally we test the algorithms in the E-learning environment, than we give the conclusion according to the results.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have