Scalable multi-view clustering with graph filtering

With the explosive growth of multi-source data, multi-view clustering has attracted great attention in recent years. Most existing methods operate raw feature space and heavily depend on quality original representation. Moreover, they are often designed for data ignore rich topology structure information. Accordingly, this paper, we propose a generic framework to cluster both attribute graph with heterogeneous features. It is capable exploring interplay between structure. Specifically, first adopt filtering technique eliminate high-frequency noise achieve clustering-friendly smooth To handle scalability challenge, develop novel sampling strategy improve anchors. Extensive experiments benchmarks demonstrate superiority our approach respect state-of-the-art approaches.