Optimizing K-Nearest Neighbor Using Grey Wolf Optimizer for Breast Cancer Classification

Breast cancer continues to be a significant global health challenge, emphasizing the need for precise methods that support early detection. This research introduces an enhanced classification framework that integrates the K-Nearest Neighbors (KNN) algorithm with the Grey Wolf Optimizer (GWO). In this approach, GWO autonomously identifies the most informative features and determines the optimal KNN parameter settings, contributing to improved model performance. The Wisconsin Diagnostic Breast Cancer dataset was utilized, and an initial exploratory analysis was conducted to better understand feature patterns and class distributions. To examine the benefit of optimization, the proposed KNN-GWO model was compared with a Principal Component Analysis (PCA) based KNN model that reduces data dimensionality. Experimental findings show that the KNN-GWO approach achieved an accuracy of 97.07%, surpassing the KNN-PCA model’s accuracy of 95.47%. The optimized model also delivered higher sensitivity and reduced false-positive predictions, both of which are crucial for clinical assessment. These results demonstrate that GWO effectively strengthens the performance of KNN while preserving the model’s interpretability and computational simplicity. Overall, this research highlights the promise of optimization-enhanced KNN techniques as dependable and transparent tools for detecting breast cancer at an early stage.