readme

-------------------Supporting code in Matlab-----------------------------------

Paper: A Nonlinear Orthogonal Non-Negative Matrix Factorization Approach to Subspace Clustering
Authors: Dijana Tolic, Nino Antulov-Fantulin, Ivica Kopriva
Contact: dijana.tolic@irb.hr
Date: 28.03.2017.
-------------------------------------------------------------------------------

Content of folder:
------------------
demo.m - runs KNSC-Ncut algorithm on synthetic data and visualizes clustering results

KNSC_NCUT.m - Kernel non-negative spectral clustering Ncut algorithm
KNSC_RCUT.m - Kernel non-negative spectral clustering Rcut algorithm
KOGNMF.m - Kernel-based orthogonal graph regularized non-negative matrix factorization

gausskernel.m - Calculates Gaussian kernel matrix
accuracy.m - Calculates the clustering accuracy with Hungarian algorithm
hungarian.m - Implementation of Hungarian algorithm

synthetic_data_6.mat - Synthetic data for demo.m