Editorial: Journal of Classification Vol. 42-1

The 42nd volume of the Journal of Classification is a treasure trove of innovative methodologies and applications in cluster analysis. The volume kicks off with an introduction to normalized clustering accuracy, a novel approach for performance assessment in cluster analysis. It then delves into the development of a finite mixture of contaminated unimodal Dirichlet distributions for clustering compositional data, offering a robust framework for handling complex datasets. The cluster validation framework based on a projection similar to Fisher’s linear discriminant analysis provides a fresh perspective on validating clustering results. The volume also addresses the critical problem of Krylov subspace solver selection, contributing to the optimization of iterative solvers. Minimum spanning tree-based methods for clustering are explored, providing insights into efficient and effective clustering techniques. A sparse mixture of generalized hyperbolic distributions is introduced for model-based clustering of high-dimensional data, showcasing advancements in handling large-scale datasets. An effective crow search algorithm for clustering is presented, offering a novel approach to optimization problems in clustering. The volume also introduces a slope stability classification model based on single-valued neutrosophic matrix energy, marking a significant step in slope stability analysis. A learning framework that combines semi-supervised clustering and classification is presented, highlighting the integration of different learning paradigms. A clustering-based oversampling algorithm for multi-class imbalance learning is proposed, addressing the challenges of imbalanced datasets. The volume extends the generalized Huber loss support vector machine from single-task to multi-task learning, demonstrating advancements in machine learning techniques. Finally, an intriguing approach that measures the distance between researchers based on their scientific output is developed, offering a unique perspective on academic research and collaboration.