AbstractFigures from this paperAbstractFigures from this paperAbstractFigures from this paperAbstractFigures from this paper

The application of intelligent optimization algorithms in feature selection has be- come an important research direction in the fields of machine learning and data mining. Feature selection aims to identify the subset of features that contribute most to model performance from high-dimensional data, in order to reduce computational complexity, eliminate redundant information, and improve model generalization. Intelligent optimization algorithms, by simulating the optimiza- tion mechanisms in nature, can efficiently explore the global optimal solution in complex search spaces. These algorithms typically combine the objective function to dynamically evaluate the quality of feature subsets and gradually approach the optimal feature combination through iterative optimization. The population initialization of the Black-winged Kite Optimization Algorithm is improved using the Tent chaotic map, and the migration and predation position updates of the Black-winged Kite are carried out through Levy flights. Addition- ally, the Golden Sine Algorithm strategy can also be used to obtain the optimal solution for the Black-winged Kite position update. Experiments show that the improved Black-winged Kite Optimization Algorithm demonstrates strong robustness and adaptability in feature selection tasks, making it particularly suitable for large-scale, high-dimensional, and nonlinear data scenarios.