**4. Conclusions and future projects**

Semantic KnowCat (SKC) is a prototype developed on KnowCat to investigate solutions to information overload in ICT-based systems, using knowledge management systems as a model. SKC uses for this purpose some hidden aspects of such systems, as the residual energy of their activity, and properties of both the elements and the activities involved.

The process of the digestion of knowledge proposed seems to be able to specify latent knowledge in a knowledge management field, which may be useful to facilitate the management task fulfilled by the system, the interaction among its entities and users' access to the contents that have been processed, among other interesting applications (Moreno-Llorena, 2008; Moreno-Llorena & Alamán, 2005; Moreno-Llorena et al., 2009a, 2009b). The enrichment of the proposed content seems to provide a very powerful support for automatic exchange of knowledge among knowledge management systems opening a way to the development of the latter on the semantic Web field (Berners-Lee, 2000).

However, the threshold found in the levels of similarity to consider the similar knowledge items is low and higher values are unlikely to appear. In almost every case taken into account most of the items having similarity over 0.3 are related to each other for their contents and the ones that aren't have minor levels, although some objectively related do not reach that value. In some cases the threshold is even lower, between 0.2 and 0.3. It would be highly desirable that the level of similarity would mark more clearly the space between items with different contents and would clarify the similarity between those that have similar contents.

With all this, it is considered highly interesting to continue advancing in an open line of work, paying special attention to specification and contrast of the level of similarity, and searching integration of content analysis proposed with the one for interaction of users (Moreno-Llorena et al., 2009a) and with automatic interaction among nodes (Moreno-Llorena et al., 2009b).
