Compara mètodes

	DBSCAN	UMAP
Camp	Aprenentatge automàtic	Aprenentatge automàtic
Família	Machine learning	Machine learning
Any d'origen	1996	2018
Autor original	Ester, M., Kriegel, H.-P., Sander, J. & Xu, X.	McInnes, L.; Healy, J.; Melville, J.
Tipus	Density-based clustering algorithm	Nonlinear manifold-learning dimension reduction
Font seminal	Ester, M., Kriegel, H.-P., Sander, J. & Xu, X. (1996). A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise. Proceedings of the 2nd KDD, 226–231.	McInnes, L., Healy, J. & Melville, J. (2018). UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction. arXiv:1802.03426.
Àlies	DBSCAN Kümeleme, density-based clustering, density-based spatial clustering	UMAP (Uniform Manifold Approximation and Projection), uniform manifold approximation and projection, manifold dimension reduction
Relacionats	3	5
Resum	DBSCAN is a density-based clustering algorithm, introduced by Ester, Kriegel, Sander and Xu in 1996, that groups together points lying in dense regions and flags points in sparse regions as noise. It is effective on noisy data and on clusters of irregular, non-spherical shapes.	UMAP (Uniform Manifold Approximation and Projection) is a fast, scalable nonlinear dimension-reduction method grounded in manifold-learning theory, introduced by McInnes, Healy and Melville in 2018. It compresses high-dimensional data into a low-dimensional embedding for visualisation and downstream analysis.