Clusteval logo ClustEval clustering evaluation framework

Here you see general information about the clustering method.

Clustering Method 'Spectral Clustering' - General

Spectral Clustering

SpectralSpectral clustering computes the second largest eigenvalue of the Laplacian of the similarity matrix to decide where to partition the similarity matrix. The resulting clusters may then be re-partitioned in order to generate a hierarchy. Efficient numeric algorithms solve the underlying linear algebra problems to facilitate spectral clustering on large datasets.

  • Publication: Alexandros Karatzoglou, Alex Smola, Kurt Hornik, and Achim Zeileis. kernlab – an S4 package for kernel methods in R. Journal of Statistical Software, 11(9):1–20, 2004.

Information Value