Fast Multi-view Discrete Clustering with Anchor Graphs

Generally, the existing graph-based multi-view clustering models consists of two steps: (1) graph construction; (2) eigen-decomposition on Laplacian matrix to compute a continuous cluster assignment matrix, followed by post-processing algorithm get discrete one. However, both construction and are time-consuming, two-stage process may deviate from directly solving primal problem. To this end, we propose Fast Multi-view Discrete Clustering (FMDC) with anchor graphs, focusing spectral problem small time cost. We efficiently generate representative anchors construct graphs different views. The is obtained performing automatically aggregated graph. FMDC has linear computational complexity respect data scale, which significant improvement compared quadratic Extensive experiments benchmark datasets demonstrate its efficiency effectiveness.