Multi-View and Multi-Order Graph Clustering Via Constrained  L1,2-Norm

Abstract

The graph-based multi-view clustering algorithms achieve decent clustering performance by consensus graph learning of the first-order graphs from different views. However, the first-order graphs are often sparse, lacking essential must-link information, which leads to suboptimal consensus graph. While high-order graphs can address this issue, a two-step strategy involving the selection of a fixed number of high-order graphs followed by their fusion may result in information loss or redundancy, restricting the exploration of high-order information. Moreover, the involvement of graphs from the views where noise outweighs useful information in learning consensus graph, results in a decline in clustering performance instead of improving clustering accuracy. So not all views are suitable for graph clustering. To address these challenges, we propose Multi-view and Multi-order Graph Clustering via Constrained L1,2-norm (MoMvGC), which mitigates the impact of graph sparsity on multi-view clustering. By introducing constrained L1,2-norm, the model ingeniously integrates the selection of multi-order graphs and corresponding weight learning into a unified framework. Furthermore, MoMvGC not only enable sparse selection of multi-order graphs but also simultaneous selection of views. We design a complete optimization framework. Comprehensive experiments conducted on nine datasets thoroughly demonstrate the effectiveness and superiority of our model.

Keywords: graph sparsityhigh-order graphconstrained L1, 2-normmulti-view clustering