In this talk, I will present a new hybrid system (FortalFS) for the problem of feature selection in machine learning. In order to evaluate the power of this new algorithm, a number of experiments were run and the results compared to well-known feature selection filter and wrapper algorithms, such as Focus, Relief, LVF, etc. Such experiments were run over eleven datasets from UCI Repository. Results show that FortalFS outperforms all other algorithms significantly in most cases. However, it presents time-consuming performance compared to random wrappers.