K Medoids Algorithm

The k-medoids algorithm is a clustering method that aims to partition data points into groups based on their similarity to representative data points, called medoids. Current research focuses on improving the algorithm's scalability and efficiency for large datasets, exploring techniques like metaheuristic optimization (e.g., using whale optimization) and developing exact polynomial-time algorithms. These advancements address the algorithm's computational limitations, making it more applicable to real-world problems in diverse fields such as time series analysis and data stream clustering, where efficient and accurate clustering is crucial. Improved seeding strategies and handling of imbalanced datasets are also active areas of investigation.