Clustering Algorithm
Clustering algorithms aim to group similar data points together, revealing underlying structure and patterns within datasets. Current research emphasizes improving scalability and efficiency for large datasets, exploring novel approaches like those based on graph neural networks, cellular automata, and dimensionality reduction techniques such as UMAP, often combined with established methods like k-means and DBSCAN. These advancements are crucial for diverse applications, ranging from image analysis and text processing to high-energy physics and urban planning, enabling more effective data exploration and knowledge discovery.
Papers
Kernel KMeans clustering splits for end-to-end unsupervised decision trees
Louis Ohl, Pierre-Alexandre Mattei, Mickaël Leclercq, Arnaud Droit, Frédéric Precioso
An enhanced Teaching-Learning-Based Optimization (TLBO) with Grey Wolf Optimizer (GWO) for text feature selection and clustering
Mahsa Azarshab, Mohammad Fathian, Babak Amiri