Journal of Computer Applications ›› 2021, Vol. 41 ›› Issue (4): 1093-1099.DOI: 10.11772/j.issn.1001-9081.2020060828

Special Issue: 人工智能

• Artificial intelligence • Previous Articles     Next Articles

Robust multi-view clustering algorithm based on adaptive neighborhood

LI Xingfeng1, HUANG Yuqing1, REN Zhenwen2, LI Yihong2   

  1. 1. School of Information Engineering, Southwest University of Science and Technology, Mianyang Sichuan 621010, China;
    2. School of National Defense Science and Technology, Southwest University of Science and Technology, Mianyang Sichuan 621010, China
  • Received:2020-06-16 Revised:2020-10-28 Online:2021-04-10 Published:2020-12-22
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (61673220), the Sichuan Major Science and Technology Project (2018TZDZX0002).


李杏峰1, 黄玉清1, 任珍文2, 李毅红2   

  1. 1. 西南科技大学 信息工程学院, 四川 绵阳 621010;
    2. 西南科技大学 国防科技学院, 四川 绵阳 621010
  • 通讯作者: 黄玉清
  • 作者简介:李杏峰(1993—),男,四川自贡人,硕士研究生,CCF会员,主要研究方向:多视图聚类、多核聚类;黄玉清(1962—),女,四川绵阳人,教授,博士,主要研究方向:信号特征提取与识别、图像处理、压缩感知;任珍文(1987—),男,四川绵阳人,讲师,博士,CCF会员,主要研究方向:多核聚类、多视图聚类、非负矩阵分解、压缩感知;李毅红(1998—),男,四川平昌人,主要研究方向:多视图聚类。
  • 基金资助:

Abstract: Since the existing adaptive neighborhood based multi-view clustering algorithms do not consider the noise and the loss of consensus graph information, a Robust Multi-View Graph Clustering(RMVGC) algorithm based on adaptive neighborhood was proposed. Firstly, to avoid the influence of noise and outliers on the data, the Robust Principal Component Analysis(RPCA) model was used to learn multiple clean low-rank data from the original data. Secondly, the adaptive neighborhood learning was employed to directly fuse multiple clean low-rank data to obtain a clean consensus affinity graph, thus reducing the information loss in the process of graph fusion. Experimental results demonstrate that the Normalized Mutual Informations(NMI) of the proposed algorithm RMVGC is improved by 5.2, 1.36, 27.2, 4.66 and 5.85 percentage points, respectively, compared to the current popular multi-view clustering algorithms on MRSCV1, BBCSport, COIL20, ORL and UCI digits datasets. Meanwhile, in the proposed algorithm, the local structure of data is maintained, the robustness against the original data is enhanced, the quality of affinity graph is improved, and such that the proposed algorithm has great clustering performance on multi-view datasets.

Key words: adaptive neighborhood, multi-view clustering, low-rank, robustness, consensus affinity graph learning

摘要: 针对现存的基于自适应邻域的多视图聚类算法没有考虑噪声和共识图信息损失的问题,提出一种基于自适应邻域的鲁棒多视图聚类(RMVGC)算法。首先,为了避免噪声和异常值对数据的影响,通过鲁棒主成分分析模型(RPCA)从原始数据中学习多个干净的低秩数据;其次,用自适应邻域学习直接融合多个干净的低秩数据来得到一个干净的共识关系图,从而减少图融合过程中的信息丢失。实验结果表明,所提RMVGC算法的标准化互信息(NMI)在MRSCV1、BBCSport、COIL20、ORL和UCI digits数据集上比目前流行的多视图聚类算法分别提升了5.2、1.36、27.2、4.66和5.85个百分点。同时,该算法保持了数据局部结构,增强了对原始数据的鲁棒性,提高了关系图质量,在多视图数据集上具有较好的聚类性能。

关键词: 自适应邻域, 多视图聚类, 低秩, 鲁棒, 共识关系图学习

CLC Number: