Cluster analysis
Cluster analysis or clustering is a way of comparing data by splitting it into groups of similar data points. These groups are called clusters.
There are many algorithms to put data into clusters. Clustering algorithms can use different ways of measuring similarity between data points.[1] As a result, different clustering algorithms can get different clusters on the same data.
Cluster Analysis Media
The result of a cluster analysis shown as the coloring of the squares into three clusters
The aurora borealis, or northern lights, above Bear Lake, Alaska
Image after running k-means clustering with k = 16
Flow diagram that shows a basic and generic approach to recommendation systems and how they utilize clustering
Single-linkage on Gaussian data. At 35 clusters, the biggest cluster starts fragmenting into smaller parts, while before it was still connected to the second largest due to the single-link effect.
Single-linkage on density-based clusters. 20 clusters extracted, most of which contain single elements, since linkage clustering does not have a notion of "noise".
k-means separates data into Voronoi cells, which assumes equal-sized clusters (not adequate here).
k-means cannot represent density-based clusters.
On Gaussian-distributed data, EM works well, since it uses Gaussians for modelling clusters.
Density-based clusters cannot be modeled using Gaussian distributions.
References
- ↑ Estivill-Castro, Vladimir. Why so many clustering algorithms: a position paper. ACM SIGKDD Explorations Newsletter 4 (1) (June 2002). p. 65–75. doi:10.1145/568574.568575.
Further reading
- Ezugwu, Absalom E.. A comprehensive survey of clustering algorithms: State-of-the-art machine learning applications, taxonomy, challenges, and future research prospects. Engineering Applications of Artificial Intelligence 110 (1 April 2022). p. 104743. doi:10.1016/j.engappai.2022.104743.
