DavidHofmeyr/SCPP: Projection Pursuit for Spectral Clustering

Performs projection pursuit based on spectral connectivity to find optimal subspaces for spectral clustering. Uses the method of Hofmeyr, D., Pavlidis, N., and Eckley, I. "Minimum spectral connectivity projection pursuit", Statistics and Computing, to appear. Clustering is performed recursively by repeatedly identifying subspaces for separating (some) clusters from others. All datasets included are from the UCI machine learning repository [https://archive.ics.uci.edu/ml/datasets.html](Bache and Lichman, 2013)

install.packages("remotes")
remotes::install_github("DavidHofmeyr/SCPP")