融合节点状态信息的跨社交网络用户对齐

doi:10.6040/j.issn.1672-3961.0.2021.281

摘要/Abstract

摘要：

提出一种融合节点状态信息的跨社交网络用户对齐方法, 通过网络表示捕获节点的局部特征和节点状态信息得到每个账户的嵌入向量, 计算不同账户对应表示之间的相似性发现对齐用户。在2个真实数据集上的试验结果表明, 提出的方法相对于其他方法可以对齐更多的用户。在预测不同尺度的top-k时, 提出的方法在网络结构较稠密的Twitter-Foursquare数据集上能够在top-9时对齐准确率达到50%且在稀疏且大网络数据集DM-ML上相比其他方法对齐准确率提高12.06%~36.62%;在分析F1-score时, 提出的方法能够有效提高用户对齐的性能。

关键词: 用户对齐, 社交网络, 局部特征, 节点状态, 网络嵌入

Abstract:

A cross social network user alignment method by fusing node state information was proposed. The local characteristics of nodes and node state information were captured through network representation to obtain the embedded vector of each account, and the aligned users were found by calculating the similarity between corresponding representations of different accounts. Experimental results on two real data sets showed that the proposed method could align more users than other methods. When predicting top-k of different scales, the proposed method could achieve an alignment precision of 50% at top-9 on the data set Twitter-Foursquare with dense network structure. Compared with other methods on the sparse and large network data set DM-ML, the improvement on alignment precision was 12.06%-36.62%. The analysis of F1-score also showed that the proposed method could effectively improve the performance of user alignment.

Key words: user alignment, social network, local characteristics, node states, network embedding

中图分类号:

TP391

胡军,杨冬梅,刘立,钟福金. 融合节点状态信息的跨社交网络用户对齐[J]. 山东大学学报 (工学版), 2021, 51(6): 49-58.

Jun HU,Dongmei YANG,Li LIU,Fujin ZHONG. Cross social network user alignment via fusing node state information[J]. Journal of Shandong University(Engineering Science), 2021, 51(6): 49-58.

图/表 7

图1

表1

表2

表3

图2

图3

图4

参考文献 26

1	DU Xingbo, YAN Junchi, ZHA Hongyuan. Joint link prediction and network alignment via cross-graph embedding[C]//Proceedings of the 28th International Joint Conference on Artificial Intelligence. Macao, China: IJCAI, 2019: 2251-2257.
2	LI Yongjun , PENG You , ZHANG Zhen , et al. Matching user accounts across social networks based on username and display name[J]. World Wide Web-Internet and Web Information Systems, 2019, 22 (3): 1075- 1097.
3	陈鸿昶, 徐乾, 黄瑞阳, 等. 一种基于用户轨迹的跨社交网络用户身份识别算法[J]. 电子与信息学报, 2018, 40 (11): 2758- 2764.
	CHEN Hongchang , XU Gan , HUANG Ruiyang , et al. A cross-social network user identification algorithm based on user trajectory[J]. Journal of Electronics and Information Technology, 2018, 40 (11): 2758- 2764.
4	XIE Wei, MU Xin, LEE K W, et al. Unsupervised user identity linkage via factoid embedding[C]//Proceedings of the 18th IEEE International Conference on Data Mining. Singapore: IEEE, 2019: 1338-1343.
5	ZHANG Si, TONG Hongang. FINAL: fast attributed network alignment[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco, USA: ACM, 2016: 1345-1354.
6	LI Yongjun , SU Zhaoting , YANG Jiaqi , et al. Exploiting similarities of user friendship networks across social networks for user identification[J]. Information Sciences, 2020, 506, 78- 98. doi: 10.1016/j.ins.2019.08.022
7	KONG Xiangnan, ZHANG Jiawei, PHILIP S Y. Inferring anchor links across multiple heterogeneous social networks[C]//Proceedings of the 22nd ACM International Conference on Information & Knowledge Management. San Francisco, USA: ACM, 2013: 179-188.
8	ZHOU Xiaoping , LIANG Xun , ZHANG Haiyan , et al. Cross-platform identification of anonymous identical users in multiple social media networks[J]. IEEE Transactions on Knowledge and Data Engineering, 2016, 28 (2): 411- 424. doi: 10.1109/TKDE.2015.2485222
9	ZHANG Jiawei, PHILIP S Y. Integrated anchor and social link predictions across social networks[C]//Proceedings of the 24th International Conference on Artificial Intelligence. Buenos Aires, Argentina: AAAI, 2015: 2125-2132.
10	ZANG Jiawei, PHILIP S Y. PCT: partial co-alignment of social networks[C]//Proceedings of the 25th International Conference World Wide Web. Montreal, Canada: ACM, 2016: 749-759.
11	SUN Song, LI Qiudan, YAN Peng, et al. Mapping users across social media platforms by integrating text and structure information[C]//IEEE International Conference on Intelligence and Security Informatics (ISI). Beijing: IEEE, 2017: 113-118.
12	MA Jiangtao , QIAO Yaqiong , HU Guangwu , et al. Balancing user profile and social network structure for anchor link inferring across multiple online social networks[J]. IEEE Access, 2017, (5): 12031- 12040.
13	LIU Li, CHUENG W K, LI Xin, et al. Aligning users across social networks using network embedding[C]//Proceedings of the 25th International Joint Conference on Artificial Intelligence. New York, USA: AAAI, 2016: 1774-1780.
14	MAN Tong, SHEN Huawei, LIU Shenghua, et al. Predict anchor links across social networks via an embedding approach[C]//Proceedings of the 25th International Joint Conference on Artificial Intelligence. New York, USA: AAAI, 2016: 1823-1829.
15	LI Xiang, SU Yijun, GAO Neng, et al. Anchor user oriented accordant embedding for user identity linkage[C]//Neural Information Processing-26th International Conference. Sydney, Australia: Springer, 2019: 561-572.
16	ZHANG Jing, CHEN Bo, WANG Xianming, et al. Mego2vec: embedding matched ego networks for user alignment across social networks[C]//Proceedings of the 27th ACM International Conference on Information and Knowledge Management. Torino, Italy: ACM, 2018: 327-336.
17	俞冬明, 李苑, 李智星, 等. 一种基于用户属性与结构的无监督用户对齐方法[J]. 南京大学学报(自然科学版), 2020, 20 (1): 1- 8.
	YU Dongming , LI Yuan , LI Zhixing , et al. An unsupervised user alignment method based on user attributes and structures[J]. Journal of Nanjing University (Natural Science Edition), 2020, 20 (1): 1- 8.
18	CHENG Anfeng, LIU Chunyin, ZHOU Chuan, et al. User alignment via structural interaction and propagation[C]//International Joint Conference on Neural Networks. Rio de Janeiro, Brazil: IEEE, 2018: 1-8.
19	FENG Shuo, WANG Qian, SHEN Derong, et al. User identification across social networks based on global view features[C]//Proceedings of the 14th Web Information Systems and Applications Conference. Liuzhou: IEEE, 2017: 93-98.
20	ZHANG Zhongbao , GU Qihang , YUE Tong , et al. Identifying the same person across two similar social networks in a unified way: globally and locally[J]. Information Sciences, 2017, 394, 53- 67.
21	TAN Shulong, GUAN Ziyu, CAI Deng, et al. Mapping users across networks by manifold alignment on hypergraph[C]//Proceedings of the 28th AAAI Conference on Artificial Intelligence (AAAI). Quebec, Canada: AAAI, 2014: 159-165.
22	ZHOU Fan, LIU Lei, ZHANG Kunpeng. et al. Deeplink: a deep learning approach for user identity linkage[C]//IEEE Conference on Computer Communications. Honolulu, USA: IEEE, 2018: 1313-1321.
23	PAGE L, BRIN S, Motwani R, et al. The pagerank citation ranking: bringing order to the web[R]. California, USA: Stanford University, 1999.
24	ZHANG Yutao, TANG Jie, YANG Zhilin, et al. COSNET: connecting heterogeneous social networks with local and global consistency[C]//Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Sydney, Australia: ACM, 2015: 1485-1494.
25	ADITYA G, JURE L. Node2vec: scalable feature learning for networks[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco, USA: ACM, 2016: 855-864.
26	CHU Xiaokai, FAN Xinxin, YAO Di, et al. Cross-network embedding for multi-network alignment[C]//The World Wide Web Conference. San Francisco, USA: ACM, 2019: 273-284.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

数据集	数据子集	节点数	边数	锚节点数
DM-ML	DBLP-Data mining	11 526	47 326	1295
DM-ML	DBLP-Machine learning	12 311	43 948	1295
Twitter-Foursquare	Twitter	5120	164 919	1609
Twitter-Foursquare	Foursquare	5313	76 972	1609

k	UAFNS	IONE	DeepLink	Mego2Vec	PALE	CrossMNA
1	0.274 7	0.205 6	0.159 4	0.145 6	0.101 2	0.015 5
5	0.435 1	0.357 5	0.312 6	0.227 8	0.208 8	0.093 2
9	0.501 6	0.439 5	0.449 3	0.291 1	0.272 1	0.183 2
17	0.564 9	0.541 1	0.551 3	0.411 4	0.360 7	0.257 7
21	0.590 4	0.566 4	0.566 8	0.436 7	0.386 0	0.291 9
25	0.604 1	0.572 7	0.584 8	0.468 4	0.411 3	0.338 5
30	0.624 0	0.601 2	0.597 3	0.481 0	0.436 7	0.363 3

k	UAFNS	IONE	DeepLink	Mego2Vec	PALE	CrossMNA
1	0.190 0	0.146 1	0.183 8	0	0.044 1	0.019 2
5	0.377 6	0.220 8	0.367 6	0.038 0	0.279 4	0.223 0
9	0.452 5	0.283 4	0.422 8	0.050 6	0.338 2	0.307 6
17	0.525 1	0.354 4	0.477 9	0.132 9	0.375 0	0.392 3
21	0.543 6	0.378 4	0.489 0	0.202 5	0.382 4	0.423 0
25	0.559 1	0.398 5	0.503 7	0.240 5	0.382 4	0.434 6
30	0.581 5	0.420 8	0.511 0	0.253 2	0.389 7	0.450 0

[1]	何奕江,杜军平,寇菲菲,梁美玉,王巍,罗盎. 基于深度卷积神经网络的图像自编码算法[J]. 山东大学学报 (工学版), 2019, 49(2): 61-66.
[2]	张东波,寇涛,许海霞. 基于LDB描述子和局部空间结构匹配的快速场景辨识[J]. 山东大学学报 (工学版), 2018, 48(5): 16-23.
[3]	读习习,刘华锋,景丽萍. 一种融合社交网络的叠加联合聚类推荐模型[J]. 山东大学学报(工学版), 2018, 48(3): 96-102.
[4]	李朔,石宇良. 基于位置社交网络中地点聚类推荐方法[J]. 山东大学学报(工学版), 2016, 46(3): 44-50.
[5]	周凯,元昌安,覃晓,郑彦,冯文铎. 基于核贝叶斯压缩感知的人脸识别[J]. 山东大学学报(工学版), 2016, 46(3): 74-78.