Compactness-Weighted KNN Classification Algorithm

The K-Nearest Neighbor (KNN) algorithm is a widely used classical classification tool, yet enhancing the classifi-cation accuracy for multi-feature large datasets remains a chal-lenge. The paper introduces a Compactness-Weighted KNN classification algorithm using a weighted Minkowski distance (CKNN) to address this. Due to the variability in sample distribu-tion, a method for deriving feature weights based on compactness is designed. Subsequently, a formula for calculating the weighted Minkowski distance using compactness weights is proposed, forming the basis for developing the CKNN algorithm. Compara-tive experimental results on five real-world datasets demonstrate that the CKNN algorithm outperforms eight existing variant KNN algorithms in Accuracy, Precision, Recall, and F1 perfor-mance metrics. The test results and sensitivity analysis confirm the CKNN's efficacy in classifying multi-feature datasets. © (2024), (Science and Information Organization). All rights reserved.