As a consequence of the natural fragmentation of science into specialities, disjoint but logically related literatures exist. Literature Based Discovery (LBD) is the science of making more evident these connections. A particular problem that scientist face today is that of the evaluation of LBD system, which is very complex. In this article we consider some of the most important aspects that should be taken into account when evaluating an LBD system. We also present a preliminary exploration of the concept of relevance in this context.