The paper introduces a novel method called Cluster-wise Anchor-based Multi-view Clustering (CAMVC) for multi-view clustering (MVC). CAMVC addresses the limitations of existing anchor-based MVC methods by focusing on learning discriminative anchors with a cluster structure regularization. The key contributions of CAMVC are: 1. **Cluster-wise Anchor Learning**: CAMVC adaptively learns multi-view anchors and a consensus subspace representation in a unified optimization model. It assumes that the latent multi-view anchors have a consensus cluster structure that matches the original data's cluster structure. 2. **Cluster Structure Regularization**: An explicit cluster structure of latent anchors is enforced by learning diverse cluster centroids, ensuring both within-cluster consistency and between-cluster diversity. 3. **Efficiency and Scalability**: The method is designed to be efficient and scalable, with a time complexity of \(O(n)\) and linear space complexity, making it suitable for large-scale datasets. The paper also includes extensive experiments on seven popular datasets, demonstrating that CAMVC outperforms state-of-the-art MVC methods in terms of clustering accuracy (ACC) and normalized mutual information (NMI). Additionally, the method's effectiveness is validated through parameter analysis, convergence analysis, and ablation studies.The paper introduces a novel method called Cluster-wise Anchor-based Multi-view Clustering (CAMVC) for multi-view clustering (MVC). CAMVC addresses the limitations of existing anchor-based MVC methods by focusing on learning discriminative anchors with a cluster structure regularization. The key contributions of CAMVC are: 1. **Cluster-wise Anchor Learning**: CAMVC adaptively learns multi-view anchors and a consensus subspace representation in a unified optimization model. It assumes that the latent multi-view anchors have a consensus cluster structure that matches the original data's cluster structure. 2. **Cluster Structure Regularization**: An explicit cluster structure of latent anchors is enforced by learning diverse cluster centroids, ensuring both within-cluster consistency and between-cluster diversity. 3. **Efficiency and Scalability**: The method is designed to be efficient and scalable, with a time complexity of \(O(n)\) and linear space complexity, making it suitable for large-scale datasets. The paper also includes extensive experiments on seven popular datasets, demonstrating that CAMVC outperforms state-of-the-art MVC methods in terms of clustering accuracy (ACC) and normalized mutual information (NMI). Additionally, the method's effectiveness is validated through parameter analysis, convergence analysis, and ablation studies.