结合FCS的多视图模糊聚类算法

doi:10.19665/j.issn1001-2400.2019.04.014

摘要/Abstract

摘要：

多视图模糊聚类综合了数据的不同表示,虽然能够产生更全面、宏观的聚类结果,但是容易受到噪声干扰。为了提高抵抗噪声的能力,提出了一种多视图模糊聚类算法。该算法同时继承了多视图聚类和模糊紧致性分离性聚类算法的优点,能够根据不同视图的重要性协同聚类,同时增强算法的鲁棒性。为了验证算法的有效性,选取4个多视图数据集进行了实验。实验结果表明,该算法不仅能够获得较高的聚类准确率,而且能有效地降低噪声数据对聚类结果的影响。

关键词: 多视图, 模糊聚类, 紧致性, 分离性

Abstract:

By synthesizing different representations of data, multi-view fuzzy clustering can produce more comprehensive and macroscopic clustering results. However it is vulnerable to noise. In order to improve the ability to resist noise, a multi-view fuzzy clustering algorithm is proposed which, inheriting the advantages of multi-view clustering and fuzzy compactness and separation clustering, can collaborate clustering according to the importance of different views and enhance robustness. In order to validate the effectiveness of this algorithm, four multi-view data sets are selected to carry out experiments. Experimental results show that this algorithm can not only achieve high clustering accuracy, but also effectively reduce the impact of noise data on clustering results.

Key words: multi-view, fuzzy clustering, compactness, separation

中图分类号:

TP391.1

刘永利,郭呈怡,刘静,吴岩. 结合FCS的多视图模糊聚类算法[J]. 西安电子科技大学学报, 2019, 46(4): 99-106.

LIU Yongli,GUO Chengyi,LIU Jing,WU Yan. Multi-view fuzzy clustering algorithm using FCS[J]. Journal of Xidian University, 2019, 46(4): 99-106.

图/表 9

表1

表2

表3

表4

表5

图1

图2

图3

图4

参考文献 15

[1]	BEZDEK J C, EHRLICH R, FULL W . FCM:the Fuzzy C-means Clustering Algorithm[J]. Computers and Geosciences, 1984,10(2/3):191-203.
[2]	WU K L, YU J, YANG M S . A Novel Fuzzy Clustering Algorithm Based on a Fuzzy Scatter Matrix with Optimality Tests[J]. Pattern Recognition Letters, 2005,26(5):639-652.
[3]	吴成茂, 上官若愚 . 嵌入隐马尔科夫随机场的中智模糊聚类算法[J]. 西安电子科技大学学报, 2017,44(6):109-114.
	WU Chengmao, SHANGGUAN Ruoyu . Neutrosophic Fuzzy Clustering Segmentation Algorithm Based on HMRF[J]. Journal of Xidian University, 2017,44(6):109-114.
[4]	KUMAR A, DAUMé III H. A Co-training Approach for Multi-view Spectral Clustering [C]//Proceedings of the 2011 28th International Conference on Machine Learning. New York: ACM, 2011: 393-400.
[5]	KUMAR A, RAI P, DAUMé III H. Co-regularized Multi-view Spectral Clustering [C]// Advances in Neural Information Processing Systems 24: 25th Annual Conference on Neural Information Processing Systems 2011. Red Hook: Curran Associates Inc, 2011: 1413-1421.
[6]	CARVALHO F D A T, MELO F M, LECHEVALLIER Y . A Multi-view Relational Fuzzy C-medoid Vectors Clustering Algorithm[J]. Neurocomputing, 2015,163:115-123.
[7]	刘正, 张国印, 陈志远 . 基于特征加权和非负矩阵分解的多视角聚类算法[J]. 电子学报, 2016,44(3):535-540. doi: 10.3969/j.issn.0372-2112.2016.03.006
	LIU Zheng, ZHANG Guoyin, CEHN Zhiyuan . A Multiview Clustering Algorithm Based on Feature Weighting and Non-negative Matrix Factorization[J]. Acta Electronica Sinica, 2016,44(3):535-540. doi: 10.3969/j.issn.0372-2112.2016.03.006
[8]	ZHANG X C, ZHAO L, ZONG L L. et al. Multi-view Clustering via Multi-manifold Regularized Nonnegative Matrix Factorization [C]//Proceedings of the 2014 IEEE International Conference on Data Mining. Piscataway: IEEE, 2014: 1103-1108.
[9]	HUANG S, KANG Z, XU Z . Self-weighted Multi-view Clustering with Soft Capped Norm[J]. Knowledge-Based Systems, 2018,158:1-8.
[10]	HUANG S, REN Y, XU Z . Robust Multi-view Data Clustering with Multi-view Capped-norm K-means[J]. Neurocomputing, 2018,311:197-208.
[11]	CAI X, NIE F, HUANG H. Multi-view K-means Clustering on Big Data[C]//Proceedings of the 2013 International Joint Conference on Artificial Intelligence. Palo Alto: AAAI Press, 2013: 2598-2604.
[12]	JIANG Y, CHUNG F L, WANG S , et al. Collaborative Fuzzy Clustering from Multiple Weighted Views[J]. IEEE Transactions on Cybernetics, 2015,45(4):688-701.
[13]	XU Y M, WANG C D, LAI J H . Weighted Multi-view Clustering with Feature Selection[J]. Pattern Recognition, 2016,53:25-35.
[14]	WANG Y, CHEN L . Multi-view Fuzzy Clustering with Minimax Optimization for Effective Clustering of Data from Multiple Sources[J]. Expert Systems with Applications, 2017,72:457-466.
[15]	JIANG B, QIU F, WANG L , et al. Bi-level Weighted Multi-view Clustering via Hybrid Particle Swarm Optimization[J]. Information Processing and Management, 2016,52(3):387-398.

算法	无噪声			有噪声
算法	Accuracy	F-Measure	NMI	Accuracy	F-Measure	NMI
FCS 1	0.314 2	0.270 8	0.244 4	0.347 6	0.282 0	0.218 6
FCS 2	0.580 9	0.564 5	0.638 4	0.595 2	0.524 3	0.681 8
FCS	0.600 0	0.580 0	0.611 2	0.580 9	0.562 8	0.626 6
RMKMC	0.623 8	0.611 0	0.593 0	0.580 9	0.576 7	0.562 8
MinimaxFCM	0.642 8	0.645 6	0.642 2	0.571 4	0.556 6	0.629 0
MvFCS	0.661 9	0.648 1	0.649 2	0.666 6	0.668 8	0.659 1

算法	无噪声			有噪声
算法	Accuracy	F-Measure	NMI	Accuracy	F-Measure	NMI
FCS 1	0.550 5	0.530 9	0.549 3	0.448 0	0.416 0	0.457 5
FCS 2	0.748 0	0.746 1	0.730 3	0.733 0	0.731 3	0.692 8
FCS	0.761 0	0.748 2	0.762 6	0.788 5	0.781 1	0.775 0
RMKMC	0.804 0	0.792 0	0.774 2	0.770 0	0.758 1	0.791 5
MinimaxFCM	0.912 0	0.912 1	0.836 4	0.801 5	0.787 1	0.786 5
MvFCS	0.913 0	0.913 1	0.838 1	0.900 0	0.900 6	0.828 8

算法	无噪声			有噪声
算法	Accuracy	F-Measure	NMI	Accuracy	F-Measure	NMI
FCS 1	0.476 1	0.455 6	0.444 4	0.428 5	0.409 8	0.440 3
FCS 2	0.683 4	0.684 3	0.553 5	0.619 0	0.603 9	0.585 6
FCS	0.789 9	0.764 1	0.848 2	0.773 1	0.748 5	0.818 8
RMKMC	0.938 3	0.848 2	0.884 2	0.778 7	0.749 8	0.783 6
MinimaxFCM	0.941 1	0.862 1	0.893 6	0.770 3	0.734 3	0.798 9
MvFCS	0.952 3	0.947 0	0.901 8	0.949 5	0.944 4	0.899 2

算法	无噪声			有噪声
算法	Accuracy	F-Measure	NMI	Accuracy	F-Measure	NMI
FCS 1	0.325 0	0.282 4	0.125 2	0.317 2	0.276 7	0.124 6
FCS 2	0.747 7	0.747 1	0.598 0	0.745 1	0.747 4	0.601 7
FCS	0.724 3	0.727 5	0.562 9	0.724 3	0.650 4	0.562 9
RMKMC	0.690 5	0.692 4	0.543 8	0.625 4	0.624 3	0.550 2
MinimaxFCM	0.725 6	0.728 7	0.568 3	0.648 8	0.649 4	0.579 4
MvFCS	0.728 2	0.731 3	0.570 5	0.672 3	0.672 8	0.585 6