ABNE: An Attention-Based Network Embedding for User Alignment Across Social Networks

Abstract

User alignment across social networks can facilitate more information/knowledge transferring across networks and thereby benefit several applications, including social link prediction, cross-domain recommendation, and information diffusion. Several works try to learn a common subspace for networks by preserving the structural proximities, such that different contribution weights of neighbors are ignored as users were always connected by unweighted edges. In this paper, we propose an attention-based network embedding model that exploits the social structures for user alignment. In particular, two main components are contained in our model framework: a masked graph attention mechanism which tries to learn the alignment task driven attention weights by the supervision of pre-aligned user pairs, and an embedding algorithm tries to learn a common vector space by explicitly modeling the weighted contribution probabilities between follower-ships and followee-ships. With the learned weights and embeddings transferring between these two components, we construct a unified model for user embedding and alignment. Stochastic gradient descent and negative sampling are adopted for efficient learning and scalability. The extensive experiments on real-world social network data sets demonstrate the effectiveness and robustness of the proposed model compared with several state-of-the-art methods. INDEX TERMS Attention mechanism, network embedding, social network analysis, user alignment.

Directed Association-based User Identification Algorithm

“…At the same time, the KL divergence that minimizes the contribution probability and the empirical probability represents each node as a low-dimensional vector, and then the vector distance calculates the similarity of the node to determine whether the node identifies. [18] proposed a network user representation model based on attention mechanism. [19][20] are all based on the user's text attributes and the representation of user nodes implemented by the multi-view network.…”

HFUL: A Hybrid Framework for User Account Linkage across Location-Aware Social Networks

“…A novel framework called "factoid embedding" is proposed by [30], the core idea of the work is that each piece of information about a user identity describes the real identity owner, and thus distinguishes the owner from others. An attention-based network embedding model was proposed in [32], and the study contains two main components: a masked graph attention mechanism and an embedding algorithm which tries to learn a common vector space. Fu et al [34] proposed a deep multi-granularity graph embedding model DeepMGGE, which utilizes the random walk to capture the higher-order structural proximities.…”

Enhanced Network Representation Learning With Community Aware and Relational Attention

“…Inspired by these recent works, attention-based architecture is introduced into NRL. Because the network nodes in the real world often show different characteristics when interacting with other nodes [20]. GAT (Graph Attention Networks) [21] updates the node representation based on the attention value of each node in the neighbors, which can deal with multiple nodes of different degrees and parallelization calculations simultaneously.…”