Unsupervised social network embedding via adaptive specific mappings

doi:10.1007/s11704-023-2180-3

Frontiers of Computer Science

2024, Vol. 18

Issue (3): 183310 https://doi.org/10.1007/s11704-023-2180-3

本期目录

Unsupervised social network embedding via adaptive specific mappings

Youming GE¹, Cong HUANG¹, Yubao LIU^1,²(

), Sen ZHANG¹, Weiyang KONG¹

¹. School of Computer Science and Engineering, Sun Yat-Sen University, Guangzhou 510275, China
². Guangdong Key Laboratory of Big Data Analysis and Processing, Guangzhou 510006, China

全文: PDF(12149 KB) HTML

Abstract：

In this paper, we address the problem of unsuperised social network embedding, which aims to embed network nodes, including node attributes, into a latent low dimensional space. In recent methods, the fusion mechanism of node attributes and network structure has been proposed for the problem and achieved impressive prediction performance. However, the non-linear property of node attributes and network structure is not efficiently fused in existing methods, which is potentially helpful in learning a better network embedding. To this end, in this paper, we propose a novel model called ASM (Adaptive Specific Mapping) based on encoder-decoder framework. In encoder, we use the kernel mapping to capture the non-linear property of both node attributes and network structure. In particular, we adopt two feature mapping functions, namely an untrainable function for node attributes and a trainable function for network structure. By the mapping functions, we obtain the low dimensional feature vectors for node attributes and network structure, respectively. Then, we design an attention layer to combine the learning of both feature vectors and adaptively learn the node embedding. In encoder, we adopt the component of reconstruction for the training process of learning node attributes and network structure. We conducted a set of experiments on seven real-world social network datasets. The experimental results verify the effectiveness and efficiency of our method in comparison with state-of-the-art baselines.

Key words： network embedding specific kernel mapping attention mechanism

收稿日期: 2022-03-31 出版日期: 2023-04-26

Corresponding Author(s): Yubao LIU

引用本文:

. [J]. Frontiers of Computer Science, 2024, 18(3): 183310.
Youming GE, Cong HUANG, Yubao LIU, Sen ZHANG, Weiyang KONG. Unsupervised social network embedding via adaptive specific mappings. Front. Comput. Sci., 2024, 18(3): 183310.

链接本文:

https://academic.hep.com.cn/fcs/CN/10.1007/s11704-023-2180-3
https://academic.hep.com.cn/fcs/CN/Y2024/V18/I3/183310

Symbol	Definition and description
$G$	The input network
$V$	The node set of the input network
$E$	The edge set of the input network
$X$	The attributes of all nodes
$x i$	The attributes of node $v i$
$z i$	The learned representations of node $v i$
$d$	The attribute dimension
$Z T$	The feature vector for learning network structures
$Z F$	The feature vector for learning node attributes
$x i^$	The reconstruction of $x i$

Tab.1

Fig.1

Tab.2

Datasets	Metrics	LN	AE	DeepWalk	N2V	UPP-SNE	ANRL	DANE	CFANE	AM-GCN	ASM
Facebook	Macro-F1	10	0.742	0.481	0.407	0.819	0.434	0.725	0.805	0.852	0.886
		30	0.844	0.506	0.460	0.820	0.437	0.776	0.834	0.854	0.892
		50	0.858	0.543	0.470	0.823	0.439	0.803	0.852	0.863	0.901
	Micro-F1	10	0.832	0.623	0.620	0.891	0.755	0.804	0.881	0.902	0.923
		30	0.890	0.698	0.775	0.892	0.757	0.839	0.904	0.905	0.931
		50	0.903	0.722	0.782	0.894	0.777	0.865	0.909	0.911	0.937
Google	Macro-F1	10	0.775	0.511	0.358	0.625	0.883	0.747	0.786	0.802	0.890
		30	0.843	0.536	0.370	0.648	0.886	0.792	0.883	0.809	0.892
		50	0.867	0.562	0.389	0.662	0.887	0.813	0.844	0.821	0.894
	Micro-F1	10	0.804	0.532	0.537	0.657	0.897	0.831	0.809	0.811	0.902
		30	0.855	0.584	0.553	0.627	0.900	0.869	0.849	0.825	0.903
		50	0.885	0.606	0.585	0.700	0.901	0.887	0.866	0.829	0.903
Twitter	Macro-F1	10	0.688	0.531	0.367	0.578	0.563	0.543	0.578	0.742	0.793
		30	0.722	0.546	0.372	0.655	0.693	0.589	0.602	0.762	0.807
		50	0.760	0.561	0.393	0.672	0.705	0.641	0.650	0.776	0.821
	Micro-F1	10	0.705	0.544	0.493	0.609	0.622	0.652	0.612	0.751	0.796
		30	0.730	0.553	0.503	0.658	0.701	0.671	0.623	0.768	0.805
		50	0.765	0.566	0.524	0.674	0.711	0.701	0.659	0.779	0.819
Citeseer	Macro-F1	10	0.594	0.442	0.489	0.552	0.660	0.605	0.663	0.659	0.687
		30	0.633	0.505	0.503	0.569	0.679	0.649	0.687	0.682	0.694
		50	0.649	0.513	0.511	0.586	0.688	0.660	0.700	0.697	0.705
	Micro-F1	10	0.644	0.482	0.537	0.614	0.717	0.673	0.710	0.688	0.734
		30	0.680	0.555	0.569	0.631	0.724	0.716	0.725	0.708	0.738
		50	0.696	0.575	0.593	0.669	0.739	0.753	0.728	0.716	0.740
BlogCatalog	Macro-F1	10	0.853	0.261	0.662	0.250	0.538	0.762	0.859	0.754	0.904
		30	0.868	0.265	0.692	0.297	0.558	0.787	0.873	0.822	0.922
		50	0.871	0.271	0.702	0.308	0.564	0.812	0.881	0.852	0.925
	Micro-F1	10	0.859	0.263	0.601	0.253	0.633	0.815	0.901	0.763	0.904
		30	0.870	0.268	0.638	0.303	0.646	0.870	0.915	0.829	0.923
		50	0.876	0.274	0.665	0.315	0.648	0.897	0.922	0.855	0.929
Flickr	Macro-F1	10	0.505	0.201	0.312	0.641	0.703	0.687	0.751	0.713	0.758
		30	0.536	0.245	0.355	0.683	0.718	0.692	0.779	0.749	0.783
		50	0.572	0.262	0.361	0.705	0.734	0.717	0.793	0.787	0.797
	Micro-F1	10	0.522	0.209	0.322	0.651	0.721	0.702	0.764	0.735	0.769
		30	0.578	0.251	0.368	0.698	0.740	0.729	0.786	0.766	0.799
		50	0.596	0.269	0.374	0.718	0.758	0.745	0.802	0.793	0.804
Cora	Macro-F1	10	0.412	0.256	0.221	0.250	0.421	0.509	0.575	0.533	0.556
		30	0.433	0.292	0.277	0.297	0.459	0.524	0.603	0.579	0.587
		50	0.462	0.343	0.334	0.308	0.488	0.545	0.619	0.608	0.613
	Micro-F1	10	0.422	0.259	0.235	0.253	0.445	0.511	0.572	0.542	0.564
		30	0.439	0.298	0.289	0.303	0.472	0.534	0.602	0.588	0.592
		50	0.477	0.351	0.340	0.315	0.502	0.563	0.625	0.620	0.615

Tab.3

Tab.4

Tab.5

Fig.2

Fig.3

Fig.4

Fig.5

Tab.6

1	X, Wang M, Zhu D, Bo P, Cui C, Shi J Pei . AM-GCN: adaptive multi-channel graph convolutional networks. In: Proceedings of the 26th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 2020, 1243−1253
2	Pan G, Yao Y, Tong H, Xu F, Lu J. Unsupervised attributed network embedding via cross fusion. In: Proceedings of the 14th ACM International Conference on Web Search and Data Mining. 2021, 797−805
3	A, Rahimi B Recht . Random features for large-scale kernel machines. In: Proceedings of the 20th International Conference on Neural Information Processing Systems. 2007, 1177−1184
4	Zhang D, Yin J, Zhu X, Zhang C. User profile preserving social network embedding. In: Proceedings of the 26th International Joint Conference on Artificial Intelligence. 2017, 3378−3384
5	T, Mikolov I, Sutskever K, Chen G, Corrado J Dean . Distributed representations of words and phrases and their compositionality. In: Proceedings of the 27th Annual Conference on Neural Information Processing Systems. 2013, 3111−3119
6	J, McAuley J Leskovec . Learning to discover social circles in ego networks. In: Proceedings of the 25th International Conference on Neural Information Processing Systems. 2012, 539−547
7	L, Tang H Liu . Relational learning via latent social dimensions. In: Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2009, 817−826
8	Y, Bengio A, Courville P Vincent . Representation learning: a review and new perspectives. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35( 8): 1798–1828
9	B, Perozzi R, Al-Rfou S Skiena . DeepWalk: online learning of social representations. In: Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2014, 701−710
10	A, Grover J Leskovec . node2vec: scalable feature learning for networks. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2016, 855−864
11	Zhang Z, Yang H, Bu J, Zhou S, Yu P, Zhang J, Ester M, Wang C. ANRL: attributed network representation learning via deep neural networks. In: Proceedings of the 27th International Joint Conference on Artificial Intelligence. 2018, 3155−3161
12	Gao H, Huang H. Deep attributed network embedding. In: Proceedings of the 27th International Joint Conference on Artificial Intelligence. 2018, 3364−3370
13	J, Tang M, Qu M, Wang M, Zhang J, Yan Q Mei . LINE: large-scale information network embedding. In: Proceedings of the 24th International Conference on World Wide Web. 2015, 1067−1077
14	D, Wang P, Cui W Zhu . Structural deep network embedding. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2016, 1225−1234
15	Qiu J, Dong Y, Ma H, Li J, Wang K, Tang J. Network embedding as matrix factorization: unifying DeepWalk, LINE, PTE, and node2vec. In: Proceedings of the 11th ACM International Conference on Web Search and Data Mining. 2018, 459−467
16	Z, Zhou Y, Gu G Yu . Adversarial network embedding using structural similarity. Frontiers of Computer Science, 2021, 15( 1): 151603
17	J, Chen Q, Zhang X Huang . Incorporate group information to enhance network embedding. In: Proceedings of the 25th ACM International Conference on Information and Knowledge Management. 2016, 1901−1904
18	Q, Long Y, Wang L, Du G, Song Y, Jin W Lin . Hierarchical community structure preserving network embedding: a subspace approach. In: Proceedings of the 28th ACM International Conference on Information and Knowledge Management. 2019, 409−418
19	Wang X, Cui P, Wang J, Pei J, Zhu W, Yang S. Community preserving network embedding. In: Proceedings of the 31st AAAI Conference on Artificial Intelligence. 2017, 203−209
20	S, Cao W, Lu Q Xu . GraRep: learning graph representations with global structural information. In: Proceedings of the 24th ACM International Conference on Information and Knowledge Management. 2015, 891−900
21	Chen H, Perozzi B, Hu Y, Skiena S. HARP: hierarchical representation learning for networks. In: Proceedings of the 32nd AAAI Conference on Artificial Intelligence. 2018, 2127−2134
22	L F R, Ribeiro P H P, Saverese D R Figueiredo . struc2vec: learning node representations from structural identity. In: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2017, 385−394
23	O, Levy Y Goldberg . Neural word embedding as implicit matrix factorization. In: Proceedings of the 27th International Conference on Neural Information Processing Systems. 2014, 2177−2185
24	Yang C, Liu Z, Zhao D, Sun M, Chang E Y. Network representation learning with rich text information. In: Proceedings of the 24th International Conference on Artificial Intelligence. 2015, 2111−2117
25	Huang X, Li J, Hu X. Label informed attributed network embedding. In: Proceedings of the 10th ACM International Conference on Web Search and Data Mining. 2017, 731−739
26	Huang X, Li J, Hu X. Accelerated attributed network embedding. In: Proceedings of 2017 SIAM International Conference on Data Mining. 2017, 633−641
27	Pan S, Wu J, Zhu X, Zhang C, Wang Y. Tri-party deep network representation. In: Proceedings of the 25th International Joint Conference on Artificial Intelligence. 2016, 1895−1901
28	Bandyopadhyay S, Lokesh N, Murty M N. Outlier aware network embedding for attributed networks. In: Proceedings of the 33rd AAAI Conference on Artificial Intelligence. 2019, 12−19
29	Yang H, Pan S, Chen L, Zhou C, Zhang P. Low-bit quantization for attributed network representation learning. In: Proceedings of the 28th International Joint Conference on Artificial Intelligence. 2019, 4047−4053
30	T N, Kipf M Welling . Variational graph auto-encoders. 2016, arXiv preprint arXiv: 1611.07308
31	T N, Kipf M Welling . Semi-supervised classification with graph convolutional networks. In: Proceedings of the 5th International Conference on Learning Representations. 2017
32	P, Velickovic G, Cucurull A, Casanova A, Romero P, Liò Y Bengio . Graph attention networks. In: Proceedings of the 6th International Conference on Learning Representations. 2018
33	K, Xu W, Hu J, Leskovec S Jegelka . How powerful are graph neural networks? In: Proceedings of the 7th International Conference on Learning Representations. 2019
34	G, Cui J, Zhou C, Yang Z Liu . Adaptive graph encoder for attributed graph embedding. In: Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2020, 976−985
35	Yang D, Wang S, Li C, Zhang X, Li Z. From properties to links: deep network embedding on incomplete graphs. In: Proceedings of 2017 ACM on Conference on Information and Knowledge Management. 2017, 367−376
36	Cao S, Lu W, Xu Q. Deep neural networks for learning graph representations. In: Proceedings of the 30th AAAI Conference on Artificial Intelligence. 2016, 1145−1152
37	G, Zhang Y, Liu X Jin . A survey of autoencoder-based recommender systems. Frontiers of Computer Science, 2020, 14( 2): 430–450
38	Li C, Wang S, Yang D, Li Z, Yang Y, Zhang X, Zhou J. PPNE: property preserving network embedding. In: Proceedings of the 22nd International Conference on Database Systems for Advanced Applications. 2017, 163−179
39	Z, Meng S, Liang J, Fang T Xiao . Semi-supervisedly co-embedding attributed networks. In: Proceedings of the 33rd International Conference on Neural Information Processing Systems. 2019, 584
40	Jian S, Hu L, Cao L, Lu K, Gao H. Evolutionarily learning multi-aspect interactions and influences from network structure and node content. In: Proceedings of the 33rd AAAI Conference on Artificial Intelligence. 2019, 598−605
41	Cen Y, Zou X, Zhang J, Yang H, Zhou J, Tang J. Representation learning for attributed multiplex heterogeneous network. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2019, 1358−1368
42	Meng Z, Liang S, Fang J, Xiao T. Semi-supervisedly co-embedding attributed networks. In: Proceedings of the 33rd International Conference on Neural Information Processing Systems. 2019, 584
43	L, Qiao F, Zhang X, Huang K, Li E Chen . Heterogeneous-attributes enhancement deep framework for network embedding. Frontiers of Computer Science, 2021, 15( 6): 156616

[1]

FCS-22180-OF-YG_suppl_1

Download

Viewed

Full text

Abstract

Cited

Shared

Discussed