Multiple clustering algorithm based on dynamic weighted tensor distance

doi:10.11772/j.issn.1001-9081.2022101626

Journal of Computer Applications ›› 2023, Vol. 43 ›› Issue (11): 3449-3456.DOI: 10.11772/j.issn.1001-9081.2022101626

• Data science and technology • Previous Articles

Multiple clustering algorithm based on dynamic weighted tensor distance

Zhuangzhuang XUE¹, Peng LI¹^,²(), Weibei FAN¹^,², Hongjun ZHANG¹, Fanshuo MENG¹

^1.School of Computer Science，Nanjing University of Posts and Telecommunications，Nanjing Jiangsu 210023，China
^2.Institute of Network Security and Trusted Computing，Nanjing University of Posts and Telecommunications，Nanjing Jiangsu 210023，China

Received:2022-11-09 Revised:2022-12-29 Accepted:2023-01-03 Online:2023-04-12 Published:2023-11-10
Contact: Peng LI
About author:XUE Zhuangzhuang， born in 1997， M. S. candidate. His research interests include machine learning， multiple clustering algorithm.
LI Peng， born in 1979， Ph. D.， professor. His research interests include computer communication networks， wireless sensor networks， information security.
FAN Weibei， born in 1987， Ph. D.， lecturer. His research interests include parallel and distributed systems， data center network， cloud computing.
ZHANG Hongjun， born in 1985， Ph. D. candidate. His research interests include multiple clustering and parallel optimization based on tensor chain decomposition.
MENG Fanshuo， born in 1997， M. S. candidate. His research interests include tensor chain decomposition.
Supported by:
National Natural Science Foundation of China(62102194);“Six Talent Peaks” High-level Talents Project of Jiangsu Province(RJFW-111)

基于动态加权张量距离的多聚类算法

薛状状¹, 李鹏¹^,²(), 樊卫北¹^,², 张宏俊¹, 孟凡朔¹

^1.南京邮电大学计算机学院、软件学院、网络空间安全学院，南京 210023
^2.南京邮电大学网络安全与可信计算研究所，南京 210023

通讯作者: 李鹏
作者简介:薛状状（1997—），男，江苏睢宁人，硕士研究生，主要研究方向：机器学习、多聚类算法
李鹏（1979—），男，福建长汀人，教授，博士，CCF会员，主要研究方向：计算机通信网络、无线传感器网络、信息安全 lipeng@njupt.edu.cn
樊卫北（1987—），男，河南开封人，讲师，博士，CCF会员，主要研究方向：并行和分布式系统、数据中心网络、云计算
张宏俊（1985—），男，安徽皖和人，博士研究生，主要研究方向：基于张量链分解的多聚类及并行优化
孟凡朔（1997—），男，江苏徐州人，硕士研究生，主要研究方向：张量链式分解。
基金资助:
国家自然科学基金资助项目(62102194);江苏省“六大人才高峰”高层次人才项目(RJFW?111)

Abstract

Abstract:

When measuring the importance of attributes in Tensor-based Multiple Clustering algorithm （TMC）， the relevance of attribute combinations within object tensors are ignored， and the selected and unselected feature space are incompletely separated because of the fixed weight strategy under different feature space selection. For above problems， a Multiple Clustering algorithm based on Dynamic Weighted Tensor Distance （DWTD-MC） was proposed. Firstly， a self-association tensor model was constructed to improve the accuracy of attribute importance measurement of each feature space. Then， a multi-view weight tensor model was built to meet the task requirements of multiple clustering analysis by dynamic weighting strategy under different feature space selection. Finally， the dynamic weighted tensor distance was used to measure the similarity of data points， generating multiple clustering results. Simulation results on real datasets show that DWTD-MC outperforms comparative algorithms such as TMC in terms of Jaccard Index （JI）， Dunn Index （DI）， Davies-Bouldin index （DB） and Silhouette Coefficient （SC）. It can obtain high quality clustering results while maintaining low redundancy among clustering results， as well as meeting the task requirements of multiple clustering analysis.

Key words: heterogeneous data, multiple clustering, tensor, tensor distance, dynamically weighting, Cyber-Physical-Social System (CPSS)

摘要：

基于张量的多聚类算法（TMC）在衡量属性重要性时忽略了对象张量内部属性组合的关联性，而且在不同的特征空间选择下，固定权重策略导致所选与未选择特征空间没有完全分离。针对上述问题，提出一种基于动态加权张量距离（DWTD）的多聚类算法（DWTD-MC）。首先，为提升各特征空间属性重要性衡量的准确性，建立了自-关联张量模型；其次，构建多视图权重张量模型，在不同特征空间选择下通过动态加权策略满足多聚类分析的需求；最后，使用DWTD衡量数据点的相似性，生成最终的多聚类结果。在真实数据集上的仿真实验结果表明，DWTD-MC在杰卡德指数（JI）、邓恩指数（DI）、DB指数（DB）和轮廓系数（SC）评价指标上均优于TMC等对比算法，而且可以在获得较高质量的聚类结果的同时，使各聚类结果之间保持较低的冗余度，满足多聚类分析的任务需求。

关键词: 异构数据, 多聚类, 张量, 张量距离, 动态加权, 社会物理信息系统

CLC Number:

TP181

Zhuangzhuang XUE, Peng LI, Weibei FAN, Hongjun ZHANG, Fanshuo MENG. Multiple clustering algorithm based on dynamic weighted tensor distance[J]. Journal of Computer Applications, 2023, 43(11): 3449-3456.

薛状状, 李鹏, 樊卫北, 张宏俊, 孟凡朔. 基于动态加权张量距离的多聚类算法[J]. 《计算机应用》唯一官方网站, 2023, 43(11): 3449-3456.

Figures/Tables 9

References 25

1	ZHOU Y， YU F R， CHEN J， et al. Cyber-physical-social systems： a state-of-the-art survey， challenges and opportunities［J］. IEEE Communications Surveys and Tutorials， 2020， 22（ 1）： 389- 425. 10.1109/comst.2019.2959013
2	MÜLLER E， ASSENT I， GÜNNEMANN S， et al. MultiClust special issue on discovering， summarizing and using multiple clusterings［J］. Machine Learning， 2015， 98（ 1/2）： 1- 5. 10.1007/s10994-014-5445-0
3	WANG J， WANG X， YU G， et al. Discovering multiple co-clusterings with matrix factorization［J］. IEEE Transactions on Cybernetics， 2021， 51（ 7）： 3576- 3587. 10.1109/tcyb.2019.2950568
4	RODRIGUEZ A， LAIO A. Clustering by fast search and find of density peaks［J］. Science， 2014， 344（ 6191）： 1492- 1496. 10.1126/science.1242072
5	徐晓，丁世飞，孙统风，等. 基于网格筛选的大规模密度峰值聚类算法［J］. 计算机研究与发展， 2018， 55（ 11）： 2419- 2429. 10.7544/issn1000-1239.2018.20170227
	XU X， DING S F， SUN T F， et al. Large-scale density peaks clustering algorithm based on grid screening［J］. Journal of Computer Research and Development， 2018， 55（ 11）： 2419- 2429. 10.7544/issn1000-1239.2018.20170227
6	BO D， WANG X， SHI C， et al. Structural deep clustering network［C］// Proceedings of the 2020 Web Conference. Republic and Canton of Geneva： International World Wide Web Conferences Steering Committee， 2020： 1400- 1410. 10.1145/3366423.3380214
7	LIU X， JI S， GLÄNZEL W， et al. Multiview partitioning via tensor methods［J］. IEEE Transactions on Knowledge and Data Engineering， 2013， 25（ 5）： 1056- 1069. 10.1109/tkde.2012.95
8	HE T， ONG Y S， HU P. Multi-source propagation aware network clustering［J］. Neurocomputing， 2021， 453： 119- 130. 10.1016/j.neucom.2021.04.064
9	ELHAMIFAR E， VIDAL R. Sparse subspace clustering： algorithm， theory， and applications［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2013， 35（ 11）： 2765- 2781. 10.1109/tpami.2013.57
10	TZORTZIS G， LIKAS A. Kernel-based weighted multi-view clustering［C］// Proceedings of the IEEE 12th International Conference on Data Mining. Piscataway： IEEE， 2012： 675- 684. 10.1109/icdm.2012.43
11	YANG S， ZHANG L. Non-redundant multiple clustering by nonnegative matrix factorization［J］. Machine Learning， 2017， 106（ 5）： 695- 712. 10.1007/s10994-016-5601-9
12	GAN G， NG M K P. Subspace clustering using affinity propagation［J］. Pattern Recognition， 2015， 48（ 4）： 1455- 1464. 10.1016/j.patcog.2014.11.003
13	欧琦媛，祝恩. 基于压缩子空间对齐的多核聚类算法［J］. 计算机工程与科学， 2021， 43（ 10）： 1730- 1735. 10.3969/j.issn.1007-130X.2021.10.003
	OU Q Y， ZHU E. Multi-kernel clustering algorithm based on compressed subspace alignment［J］. Computer Engineering and Science， 2021， 43（ 10）： 1730- 1735. 10.3969/j.issn.1007-130X.2021.10.003
14	ZHAO Y， YANG L T， ZHANG R. A tensor-based multiple clustering approach with its applications in automation systems［J］. IEEE Transactions on Industrial Informatics， 2018， 14（ 1）： 283- 291. 10.1109/tii.2017.2748800
15	唐益明，丰刚永，任福继，等. 面向结构复杂数据集的模糊聚类有效性指标［J］. 电子测量与仪器学报， 2018， 32（ 4）： 119- 127.
	TANG Y M， FENG G Y， REN F J， et al. Fuzzy clustering validity index facing data set with complexity structure［J］. Journal of Electronic Measurement and Instrumentation， 2018， 32（ 4）： 119- 127.
16	严加展，陈华，李阳. 改进的模糊C-均值聚类有效性指标［J］. 计算机工程与应用， 2020， 56（ 9）： 156- 161. 10.3778/j.issn.1002-8331.1901-0117
	YAN J Z， CHEN H， LI Y. Improved fuzzy C-means clustering validity index［J］. Computer Engineering and Applications， 2020， 56（ 9）： 156- 161. 10.3778/j.issn.1002-8331.1901-0117
17	KOLDA T G， BADER B W. Tensor decompositions and applications［J］. SIAM Review， 2009， 51（ 3）： 455- 500. 10.1137/07070111x
18	LIU Y， LIU Y， CHAN K C C. Tensor distance based multilinear locality-preserved maximum information embedding［J］. IEEE Transactions on Neural Networks， 2010， 21（ 11）： 1848- 1854. 10.1109/tnn.2010.2066574
19	匡立伟. 基于张量的大数据统一表示及降维方法研究［D］. 武汉：华中科技大学， 2016： 32- 35.
	KUANG L W. A tensor-based approach for united representation and dimensionality reduction of big data［D］. Wuhan： Huazhong University of Science and Technology， 2016： 32- 35
20	JOSE M， MARTINEZ V， REINALDO A， et al. On the limiting probabilities of the M/E_r/1 queueing system［J］. Statistics and Probability Letters， 2014， 88： 56- 61. 10.1016/j.spl.2014.01.030
21	GUO P-C， GAO S-C， GUO X-X. A modified newton method for multilinear PageRank［J］. Journal of Mathematics， 2018， 22（ 5）： 1161- 1171. 10.11650/tjm/180303
22	KUANG L， HAO F， YANG L T， et al. A tensor-based approach for big data representation and dimensionality reduction［J］. IEEE Transactions on Emerging Topics in Computing， 2014， 2（ 3）： 280- 291. 10.1109/tetc.2014.2330516
23	ZHAO Y， YANG L T， ZHANG R. Tensor-based multiple clustering approaches for cyber-physical-social applications［J］. IEEE Transactions on Emerging Topics in Computing， 2020， 8（ 1）： 69- 81. 10.1109/tetc.2018.2801464
24	LIU S， ZHENG J， LIN Q， et al. Evolutionary multi and many-objective optimization via clustering for environmental selection［J］. Information Sciences， 2021， 578： 930- 949. 10.1016/j.ins.2021.08.054
25	ZHENG X， LIU X， CHEN J， et al. Adaptive partial graph learning and fusion for incomplete multi-view clustering［J］. International Journal of Intelligent Systems， 2022， 37（ 1）： 991- 1009. 10.1002/int.22655

空气质量特征	记录范围	空气质量特征	记录范围
PM2.5	［0，500］	O₃	［0，0.3］
SO₂	［0，0.43］	CO	［0，14］
NO₂	［0，0.18］

空气质量特征	记录范围	空气质量特征	记录范围
PM2.5	［0，500］	O₃	［0，0.3］
SO₂	［0，0.43］	CO	［0，14］
NO₂	［0，0.18］

数据集	对象数	算法	JI	DI	DB	SC
数据集1	300	DWTD-MC	0.342	0.248	0.382	0.735
		TMC	0.426	0.183	0.586	0.654
		NWTD-MC	0.494	0.166	1.022	0.463
		TD-MC	0.372	0.209	0.483	0.722
	500	DWTD-MC	0.331	0.267	0.562	0.746
		TMC	0.406	0.189	0.756	0.756
		NWTD-MC	0.487	0.149	0.965	0.362
		TD-MC	0.357	0.210	0.530	0.732
数据集2	500	DWTD-MC	0.310	0.298	0.456	0.701
		TMC	0.419	0.188	0.698	0.532
		NWTD-MC	0.576	0.170	1.154	0.465
		TD-MC	0.361	0.222	0.632	0.592
	800	DWTD-MC	0.287	0.270	0.498	0.695
		TMC	0.448	0.200	0.568	0.433
		NWTD-MC	0.579	0.165	0.984	0.235
		TD-MC	0.355	0.223	0.521	0.523

数据集	对象数	算法	JI	DI	DB	SC
数据集1	300	DWTD-MC	0.342	0.248	0.382	0.735
		TMC	0.426	0.183	0.586	0.654
		NWTD-MC	0.494	0.166	1.022	0.463
		TD-MC	0.372	0.209	0.483	0.722
	500	DWTD-MC	0.331	0.267	0.562	0.746
		TMC	0.406	0.189	0.756	0.756
		NWTD-MC	0.487	0.149	0.965	0.362
		TD-MC	0.357	0.210	0.530	0.732
数据集2	500	DWTD-MC	0.310	0.298	0.456	0.701
		TMC	0.419	0.188	0.698	0.532
		NWTD-MC	0.576	0.170	1.154	0.465
		TD-MC	0.361	0.222	0.632	0.592
	800	DWTD-MC	0.287	0.270	0.498	0.695
		TMC	0.448	0.200	0.568	0.433
		NWTD-MC	0.579	0.165	0.984	0.235
		TD-MC	0.355	0.223	0.521	0.523

[1]	Yunfei SHEN, Fei SHEN, Fang LI, Jun ZHANG. Deep neural network model acceleration method based on tensor virtual machine [J]. Journal of Computer Applications, 2023, 43(9): 2836-2844.
[2]	Shaofa SHANG, Lin JIANG, Yuancheng LI, Yun ZHU. Adaptive partitioning and scheduling method of convolutional neural network inference model on heterogeneous platforms [J]. Journal of Computer Applications, 2023, 43(9): 2828-2835.
[3]	Xinmin SHI, Yong LIU, Yaojian CHEN, Jiawei SONG, Xin LIU. Acceleration and optimization of quantum computing simulator implemented on new Sunway supercomputer [J]. Journal of Computer Applications, 2023, 43(8): 2486-2492.
[4]	Changhong LIU, Sheng ZENG, Bin ZHANG, Yong CHEN. Cross-modal tensor fusion network based on semantic relation graph for image-text retrieval [J]. Journal of Computer Applications, 2022, 42(10): 3018-3024.
[5]	Haoyun LEI, Zenwen REN, Yanlong WANG, Shuang XUE, Haoran LI. Multiple kernel clustering algorithm based on capped simplex projection graph tensor learning [J]. Journal of Computer Applications, 2021, 41(12): 3468-3474.
[6]	Jicheng CHEN, Hongchang CHEN. Community detection method based on tensor modeling and evolutionary K-means clustering [J]. Journal of Computer Applications, 2021, 41(11): 3120-3126.
[7]	CHEN Lixia, BAN Ying, WANG Xuewen. Background subtraction based on tensor nuclear norm and 3D total variation [J]. Journal of Computer Applications, 2020, 40(9): 2737-2742.
[8]	GONG Kaiqiang, ZHANG Chunmei, ZENG Guanghua. Convolution neural network model compression method based on pruning and tensor decomposition [J]. Journal of Computer Applications, 2020, 40(11): 3146-3151.
[9]	WANG Xin, GAO Yuan, WANG Bin, SUN Jie, XIANG Jie. Application of multimodal network fusion in classification of mild cognitive impairment [J]. Journal of Computer Applications, 2019, 39(12): 3703-3708.
[10]	FENG Wei, YI Mianzhu, MA Yanzhou. Russian phonetic transcription system based on TensorFlow [J]. Journal of Computer Applications, 2018, 38(4): 971-977.
[11]	REN Xiaoxu, LYU Liangfu, CUI Guangtai. Data fusion algorithm of coupled images [J]. Journal of Computer Applications, 2018, 38(10): 2862-2868.
[12]	YU Keqin, WU Yingbo, LI Shun, JIANG Jiacheng, XIANG De, WANG Tianhui. Tensor factorization recommendation algorithm based on context similarity of mobile user [J]. Journal of Computer Applications, 2017, 37(9): 2531-2535.
[13]	YAN Honghai, PU Fangling, XU Xin. Video super resolution method based on structure tensor [J]. Journal of Computer Applications, 2016, 36(7): 1944-1948.
[14]	DING Xiaohuan, PENG Furong, WANG Qiong, LU Jianfeng. Co-clustering recommendation algorithm based on parallel factorization decomposition [J]. Journal of Computer Applications, 2016, 36(6): 1594-1598.
[15]	LIN Jing, YANG Jichen, ZHANG Xueyuan, LI Xinchao. Audio feature extraction algorithm based on weight tensor of sparse representation [J]. Journal of Computer Applications, 2016, 36(5): 1426-1429.

Multiple clustering algorithm based on dynamic weighted tensor distance

基于动态加权张量距离的多聚类算法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 9

References 25

Related Articles 15

Recommended Articles

Metrics