块对角子空间聚类中成对约束的主动式学习

doi:10.6040/j.issn.1672-3961.0.2020.182

山东大学学报 (工学版) ›› 2021, Vol. 51 ›› Issue (2): 65-73.doi: 10.6040/j.issn.1672-3961.0.2020.182

• • 上一篇

块对角子空间聚类中成对约束的主动式学习

解子奇,王立宏^*,李嫚

烟台大学计算机与控制工程学院, 山东烟台 264005

发布日期:2021-04-16
作者简介:解子奇(1996— ),男,山东济南人,硕士研究生,主要研究方向为聚类分析. E-mail:xzzq1996@sina.com. *通信作者简介:王立宏(1970— ),女,吉林镇赉人,教授,博士,主要研究方向为数据挖掘与知识发现. E-mail:wanglh@ytu.edu.cn
基金资助:
国家自然科学基金资助项目(61773331,72072154)

Active learning of pairwise constraints in block diagonal subspace clustering

XIE Ziqi, WANG Lihong^*, LI Man

School of Computer and Control Engineering, Yantai University, Yantai 264005, Shandong, China

Published:2021-04-16

摘要/Abstract

摘要： 针对块对角表示(block diagonal representation, BDR)子空间聚类算法在对子空间重叠的高维数据聚类时效果较差的问题,提出成对约束的块对角子空间聚类(constrained subspace clustering with block diagonal representation, CBDR)算法,设计主动式学习策略,获取用户提供的少量数据点成对信息,以改进BDR算法的性能,给出CBDR算法的目标函数和求解过程。在测试集上的试验结果表明,CBDR算法的聚类错误率和归一化互信息指标比BDR和SBDR(structured block diagonal representation)算法好,而且主动式选取点对方法优于随机选取点对方法,使用少于5‰的约束信息可降低BDR的聚类错误率达到5%以上。

关键词: 子空间聚类, 主动式学习, 成对约束, 块对角表示, 约束聚类

Abstract: Focusing on the poor performance of subspace clustering by block diagonal representation(BDR)on high-dimensional data with overlapped subspaces, an active learning strategy was designed to obtain partial pairwise information among a few data points. A pairwise constrained block diagonal representation algorithm(CBDR)was proposed to improve the performance of the BDR algorithm. The objective function and solution process of the CBDR were given. The experimental results on the test datasets showed that the CBDR algorithm reduced the clustering error by more than 5% with less than 5‰ constraint information in terms of clustering error and normalized mutual information, which significantly outperformed the compared algorithms, i.e., BDR, SBDR(structured block diagonal representation)with random selection of pairwise constraints.

Key words: subspace clustering, active learning, pairwise constraints, block diagonal representation, constrained clustering

中图分类号:

TP181

解子奇,王立宏,李嫚. 块对角子空间聚类中成对约束的主动式学习[J]. 山东大学学报 (工学版), 2021, 51(2): 65-73.

XIE Ziqi, WANG Lihong, LI Man. Active learning of pairwise constraints in block diagonal subspace clustering[J]. Journal of Shandong University(Engineering Science), 2021, 51(2): 65-73.

参考文献

[1] ELHAMIFAR E, VIDAL R. Sparse subspace clustering: algorithm, theory, and applications[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2012, 35(11):2765-2781.
[2] LIU G, LIN Z, YAN S, et al. Robust recovery of subspace structures by low-rank representation[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2013, 35(1):171-184.
[3] LU C, FENG J, LIN Z, et al. Subspace clustering by block diagonal representation[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2019, 41(2):487-501.
[4] YANG Y, ZHANG X. Subspace clustering algorithm based on Laplacian rank constraint[C] //Proceedings of 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference. Chengdu, China: IEEE, 2019:1556-1559.
[5] VIDAL R, MA Y, SASTRY S. Generalized principal component analysis[M]. New York, USA: Springer, 2016.
[6] WANG L, HUANG J, YI M, et al. Block diagonal representation learning for robust subspace clustering[J]. Information Sciences, 2020, 526:54-67.
[7] ZHANG Zhao, REN Jiahuan, LI Sheng, et al. Robust subspace discovery by block-diagonal adaptive locality-constrained representation[C] //Proceedings of ACM International Conference on Multimedia(MM'19). Nice, France: ACM, 2019.
[8] 郑建炜, 李卓蓉, 王万良, 等. 联合Laplacian正则项和特征自适应的数据聚类算法[J]. 软件学报, 2019, 30(12):3846-3861. ZHENG Jianwei, LI Zhuorong, WANG Wanliang, et al. Clustering with joint Laplacian regularization and adaptive feature learning[J]. Journal of Software, 2019, 30(12):3846-3861.
[9] HE R, ZHANG Y, SUN Z, et al. Robust subspace clustering with complex noise[J]. IEEE Transactions on Image Processing, 2015, 24(11):4001-4013.
[10] 鲁全茂. 面向高维数据的聚类算法研究[D]. 北京:中国科学院大学, 2018. LU Quanmao. Research on clustering algorithms for high-dimensional data[D]. Beijing:University of Chinese Academy of Sciences, 2018.
[11] ABDOLALI M, RAHMATI M. Neither global nor local: a hierarchical robust subspace clustering for image data[J]. Information Sciences, 2020, 514:333-353.
[12] LIU Maoshan, WANG Yan, SUN Jun, et al. Structured block diagonal representation for subspace clustering[J]. Applied Intelligence, 2020.
[13] ZHANG Zhao, ZHANG Yan, LIU Guangcan, et al. Joint label prediction based semi-supervised adaptive concept factorization for robust data representation[J]. IEEE Transactions on Knowledge and Data Engineering, 2020, 32(5):952-970.
[14] YIN M, XIE S, WU Z, et al. Subspace clustering via learning an adaptive low-rank graph[J]. IEEE Transactions on Image Processing, 2018, 27(8):3716-3728.
[15] WANG Weiwei, YANG Chunyu, CHEN Huazhu, et al. Unified discriminative and coherent semi-supervised subspace clustering[J]. IEEE Transactions on Image Processing, 2018, 27(5):2461-2470.
[16] LIU Y, LIU K, ZHANG C, et al. Entropy-based active sparse subspace clustering[J]. Multimedia Tools and Applications, 2018, 77:22281-22297.
[17] WANG J, WANG X, TIAN F, et al. Constrained low-rank representation for robust subspace clustering[J]. IEEE Transactions on Cybernetics, 2017, 47(12):4534-4546.
[18] WAGSTAFF K, CARDIE C, ROGERS S, et al. Constrained k-means clustering with background knowledge[C] //Proceedings of the Eighteenth International Conference on Machine Learning. MA, USA: Morgan Kaufmann Publishers Inc, 2001:577-584.
[19] LUXBURG U V. A tutorial on spectral clustering[J]. Statistics & Computing, 2007, 17(4):395-416.
[20] NIE F, WANG H, CAI X, et al. Robust matrix completion via joint schatten p-norm and lp-norm minimization[C] //Proceedings of IEEE International Conference on Data Mining Series. Brussels, Belgium: IEEE, 2012:566-574.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

块对角子空间聚类中成对约束的主动式学习

Active learning of pairwise constraints in block diagonal subspace clustering

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

多维度评价

本文评价

推荐阅读 0

[1]	丁彦,李永忠*. 基于PCA和半监督聚类的入侵检测算法研究[J]. 山东大学学报(工学版), 2012, 42(5): 41-46.
[2]	张友新,王立宏. 两阶段近邻传播半监督聚类算法[J]. 山东大学学报(工学版), 2012, 42(2): 18-22.
[3]	张道强. 知识保持的嵌入方法[J]. 山东大学学报(工学版), 2010, 40(2): 1-10.