Abstract: Based on the metaheuristic algorithm： crow search algorithm (CrSA), we proposed an improved feature selection based on crow search algorithm (IFSCrSA) to solve the shortcomings of the current feature selection problem. Compared with traditional machine learning feature selection algorithms and feature selection algorithm based on evolutionary computing. The results show that IFSCrSA can select features with strong recognition in the data set, which not only greatly reduces the size of the feature subset, but also improves the classification accuracy.

Key words: metaheuristic algorithm, crow search algorithm, feature selection, classification accuracy