With the consolidation of Smart Grids, a considerable amount of works can be noticed, mainly with respect to the application of several artificial intelligence techniques in order to automatically identify non-technical losses, but the problem of selecting the most representative features has not been widely discussed. In this work, we make a parallel among the problem of non-technical losses and the task of irregular consumers characterization by means of a recent meta-heuristic optimization technique called Black Hole Algorithm (BHA). The experimental setup is conducted over two private datasets provided by a Brazilian electric power company, and it shows the importance of selecting the most relevant features in the context of non-technical losses identification, as well as the suitability of BHA to this task.