Figure 2

Outline of the implemented document classification (Part II). The outline of the implemented methodology is demonstrated via an imaginary example (continued): after the input has been entered and the defined data have been collected (see Figure 1), during the next step (2. Processing) the abstracts indicated by the user are processed to be eventually represented in the format LIBLINEAR understands. First the terms pre-determined by the user (and the standard ones) are retrieved, as well as their frequencies in each abstract (the values presented in the figure do not represent a real scenario; they are used just to indicate how the process works) - the features used in the vectors correspond only to the selected terms found in the texts of the abstracts from sets A and B - a meta-arrangement uses vectors from sets A and B both as training set and default test set.