Multiple Non-Redundant Spectral Clustering Views
Donglin Niu, Jennifer Dy, Michael I. Jordan · 2010
Many clustering algorithms only find one clustering solution. However, data can of-ten be grouped and interpreted in many dif-ferent ways. This is particularly true in the high-dimensional setting where differ-ent subspaces reveal different possible group-ings of the data. Instead of committing to one clustering solution, here we intro-duce a novel method that can provide sev-eral non-redundant clustering solutions to the user. Our approach simultaneously learns non-redundant subspaces that provide multi-ple views and finds a clustering solution in each view. We achieve this by augmenting a spectral clustering objective function to in-corporate dimensionality reduction and mul-tiple views and to penalize for redundancy between the views. 1.