The rapid growth of data and the need for its proper analysis still presents a big problem for intelligent data analysis and machine learning algorithms. In order to gain a better insight into the problem being analyzed, researchers today are trying to find solutions for reducing the dimensionality of the data, by adopting algorithms that could reveal the most informative features out of the data. For this purpose, in this paper we propose a novel feature selection method based on differential evolution with a threshold mechanism. The proposed method was tested on a phishing website classification problem and evaluated with two experiments. The experimental results revealed that the proposed method performed the best in all of the test cases.