26 February 2024 in IEEE Transactions on Multimedia
Recently, multi-view clustering methods have been widely used in handling multi-media data and have achieved impressive performances. Among the many multi-view clustering methods, anchor graph-based multi-view clustering has been proven to be highly efficient for large-scale data processing. However, most existing anchor graph-based clustering methods necessitate post-processing to obtain clustering labels and are unable to effectively utilize the information within anchor graphs. To address this issue, we draw inspiration from regression and feature selection to propose A nchor G raph-Based F eature S election for O ne-step M ulti- V iew C lustering (AGFS-OMVC). Our method combines embedding learning and sparse constraint to perform feature selection, allowing us to remove noisy anchor points and redundant connections in the anchor graph. This results in a clean anchor graph that can be projected into the label space, enabling us to obtain clustering labels in a single step without post-processing. Lastly, we employ the tensor Schatten $p$ -norm as a tensor rank approximation function to capture the complementary information between different views, ensuring similarity between cluster assignment matrices. Experimental results on five real-world datasets demonstrate that our proposed method outperforms state-of-the-art approaches.
4 September 2023 in Neural Networks
24 February 2025 in IEEE Transactions on Image Processing
Large-scale multi-view clustering for image data has achieved impressive clustering performance and efficiency. However, most methods lack interpretability in clustering and do not fully consider the complementarity of distributions between different views. To address these problems, we introduce Multi-View Clustering with Transition Probabilities Learning (MVCTPL). Specifically, we construct an anchor graph factorization model from the perspective of transition probabilities, while simultaneously learning transition probability matrices from samples to clusters and from anchor points to clusters, serving as soft label matrices for samples and anchor points, respectively. This model enables one-step label acquisition and provides the model with a sound probability interpretation. Moreover, since the clusters of samples and anchor points should be consistent across all views, we employ Schatten p-norm regularization on the two matrices, effectively mining the complementary information distributed among the views, thereby aligning the labels across views more consistently. Comprehensive testing on four small-scale datasets and three large-scale datasets confirms the effectiveness of this model.