Representing documents by vectors that are independent of language enhances machine translation and multilingual text categorization. We use discriminative training to create a pr...
A novel text extraction method from graphical document images is presented in this paper. Graphical document images containing text and graphics components are considered as two-d...
One common predictive modeling challenge occurs in text mining problems is that the training data and the operational (testing) data are drawn from different underlying distributi...
In recent years research towards Indian handwritten character recognition is getting increasing attention. Many approaches have been proposed by the researchers towards handwritte...