Clustering methods in machine learning

Clustering is a fundamental task in machine learning for discovering hidden structures in unlabelled data. The article reviews key clustering methods, including centroid, density, hierarchy and model-based approaches. Their advantages, limitations and applications are analysed to provide a comprehensive overview of the state of clustering in machine learning. Their effectiveness is compared on the basis of selected metrics to evaluate the outcome of a given clustering. Recent developments and challenges, including scalability and interpretability problems, are also discussed.