One-step Multi-view Clustering With Adaptive Low-rank Anchor-graph Learning

2025-09-18

AI Summary

Existing anchor graph-based multi-view clustering (AGMC) methods construct the consensus anchor graph (CAG) by naively fusing view-specific anchor graphs, making them vulnerable to redundancy and noise; moreover, anchor graph learning and clustering indicator matrix generation are performed separately, leading to suboptimal performance and low efficiency. To address these issues, this paper proposes an integrated AGMC framework that jointly optimizes anchor graph learning and clustering indicator matrix generation. Specifically, it employs adaptive low-rank constraints and nuclear norm regularization to suppress noise and enhance cross-view consistency, and introduces an end-to-end multi-view fusion mechanism that unifies anchor selection, CAG construction, and cluster assignment. Extensive experiments on multiple standard and large-scale benchmark datasets demonstrate that the proposed method significantly improves clustering accuracy and computational efficiency, while maintaining scalability and robustness.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Multi-modal Vision
• Multiagent Systems: Multiagent Learning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text

Abstract

In light of their capability to capture structural information while reducing computing complexity, anchor graph-based multi-view clustering (AGMC) methods have attracted considerable attention in large-scale clustering problems. Nevertheless, existing AGMC methods still face the following two issues: 1) They directly embedded diverse anchor graphs into a consensus anchor graph (CAG), and hence ignore redundant information and numerous noises contained in these anchor graphs, leading to a decrease in clustering effectiveness; 2) They drop effectiveness and efficiency due to independent post-processing to acquire clustering indicators. To overcome the aforementioned issues, we deliver a novel one-step multi-view clustering method with adaptive low-rank anchor-graph learning (OMCAL). To construct a high-quality CAG, OMCAL provides a nuclear norm-based adaptive CAG learning model against information redundancy and noise interference. Then, to boost clustering effectiveness and efficiency substantially, we incorporate category indicator acquisition and CAG learning into a unified framework. Numerous studies conducted on ordinary and large-scale datasets indicate that OMCAL outperforms existing state-of-the-art methods in terms of clustering effectiveness and efficiency.

Problem

• Improves multi-view clustering by reducing redundant information and noise in anchor graphs
• Integrates category indicator learning and consensus graph construction into unified framework
• Enhances both clustering effectiveness and efficiency for large-scale datasets

Innovation

• Adaptive low-rank anchor-graph learning
• Nuclear norm-based consensus graph model
• Unified framework for indicator and graph learning