General overview of data stream clustering

Abstract

Data stream clustering is a widely researched problem in data mining and machine learning. This involves grouping data points that arrive sequentially and continuously over time, making it difficult to work with because of the large amounts of data that need to be processed in real-time. This work reviews recent advances in data stream clustering algorithms and techniques, including traditional and online clustering methods, ensemble methods, and stream-based outlier detection, discusses the evaluation metrics for data stream clustering, and highlights some open research challenges in this area.