Genetic datasets have a large number of features that may significantly affect the disease classification process, especially datasets related to cancer diseases. Evolutionary algorithms (EA) are used to find the fastest and best way to perform these calculations, such as the bat algorithm (BA) by reducing the dimensions of the search area after changing it from continuous to discrete. In this paper, a method of gene selection was proposed two sequent stages: in the first stage, the fuzzy mutual information (FMI) method is used to choose the most important genes selected through a fuzzy model that was built based on the dataset size. In the second stage, the BBA is used to reduce and determine a fixed number of genes affecting the process of classification, which came from the first stage. The proposed algorithm, FMI_BBA, describes efficiency, by obtaining a higher classification accuracy and a few numbers of selected genes compared to other algorithms.