The paper introduces a new binary version of the Spider Wasp Optimizer (BSWO) called Binary Boosted SWO (BBSWO) to address the challenging task of High-dimensional Feature Selection (HFS). The BBSWO combines several successful strategies to enhance the performance of HFS, addressing the limitations of the original BSWO, such as early convergence, local optimum settling, limited exploration and exploitation, and lack of population diversity. The BBSWO incorporates chaos optimization using sine chaos mapping for initialization, a new convergence parameter for balancing exploration and exploitation, multiple exploration mechanisms, and quantum-based optimization to improve search agent diversity. The proposed method was evaluated using the k-Nearest Neighbor (k-NN) classifier on 23 HFS problems from the biomedical domain, showing superior performance compared to traditional BSWO and other meta-heuristic-based feature selection methods. The results indicate that BBSWO can efficiently identify the least significant feature subsets with high classification accuracy.The paper introduces a new binary version of the Spider Wasp Optimizer (BSWO) called Binary Boosted SWO (BBSWO) to address the challenging task of High-dimensional Feature Selection (HFS). The BBSWO combines several successful strategies to enhance the performance of HFS, addressing the limitations of the original BSWO, such as early convergence, local optimum settling, limited exploration and exploitation, and lack of population diversity. The BBSWO incorporates chaos optimization using sine chaos mapping for initialization, a new convergence parameter for balancing exploration and exploitation, multiple exploration mechanisms, and quantum-based optimization to improve search agent diversity. The proposed method was evaluated using the k-Nearest Neighbor (k-NN) classifier on 23 HFS problems from the biomedical domain, showing superior performance compared to traditional BSWO and other meta-heuristic-based feature selection methods. The results indicate that BBSWO can efficiently identify the least significant feature subsets with high classification accuracy.