Analysis of the performance of algorithms (K-Means, Farthest First, Hierarchical) Using the data analysis and modeling tool Weka

Abstract

This study assesses the performance of three clustering algorithms—K-Means, Farthest First, and Hierarchical—using the Weka data mining tool. These algorithms were applied to five diverse datasets representing healthcare, industrial, and benchmark applications to evaluate their clustering accuracy, execution time, and consistency. The experimental results show that the Farthest First algorithm achieves the highest accuracy and the fastest execution time, making it suitable for real-time applications. K-Means delivers balanced performance but is sensitive to initialization and outliers, while the Hierarchical algorithm effectively captures complex relationships but incurs high computational costs. The findings highlight the importance of selecting appropriate clustering techniques based on dataset characteristics and application requirements. Future work will explore advanced clustering methods such as DBSCAN and Gaussian Mixture Models to improve scalability and performance on large datasets.