An L1-Version of the Spectral Clustering
Compute the performances of the l1-spectral clustering algorithm
Create data set
Find the representative elements of the clusters
Find the optimal number of clusters
Find the structure of the graph from the adjacency matrix
Run the l1-spectral clustering algorithm on one component
Run the l1-spectral clustering algorithm
Description of the package
Solve the internal minimization problem
Provides an l1-version of the spectral clustering algorithm devoted to robustly clustering highly perturbed graphs using l1-penalty. This algorithm is described with more details in the preprint C. Champion, M. Champion, M. Blazère, R. Burcelin and J.M. Loubes, "l1-spectral clustering algorithm: a spectral clustering method using l1-regularization" (2022).