GALA: Group Merging Artificial Lemming Algorithm for Gene Selection

Feature selection aims to identify the optimal feature subset from high-dimensional data to improve the performance of the model. However, traditional feature selection methods often show limitations when applied to complex high-dimensional and small-sample gene sequencing datasets and often fail to obtain satisfactory modeling results. In this paper, the t-test is adopted to select 100 features. The feature set is divided into two groups equally according to the sorting. For each group, the artificial lemming algorithm (ALA) is used for feature selection. The two sets of important information subsets are merged, and the ALA algorithm is continued to be used to further refine the feature set. The experimental results on seven high-dimensional datasets prove the effectiveness and efficiency of this method, which is superior to the existing methods.