Exploiting discriminant and SVM constraints in NMF

A novel method is introduced for exploiting the support vector machine and additional discriminant constraints in nonnegative matrix factorization. The notion of the proposed method is to find the projection matrix that projects the data to a low-dimensional space so that the data projections have minimum within-class variance, maximum between-class variance and the data projections between the two classes are separated by a hyperplane with maximum margin. Experiments were performed on several two-class UCI data sets, as well as on the Cohn-Kanade database for facial expression recognition. Experimental results showed that the proposed method achieves better classification performance than the state of the art nonnegative matrix factorization and discriminant nonnegative matrix factorization followed by support vector machines classification.