Abstract:Recently, Network Embedding (NE) has become one of the most attractive research topics in machine learning and data mining. NE approaches have achieved promising performance in various of graph mining tasks including link prediction and node clustering and classification. A wide variety of NE methods focus on the proximity of networks. They learn community-oriented embedding for each node, where the corresponding representations are similar if two nodes are closer to each other in the network. Meanwhile, there is another type of structural similarity, i.e., role-based similarity, which is usually complementary and completely different from the proximity. In order to preserve the role-based structural similarity, the problem of role-oriented NE is raised. However, compared to community-oriented NE problem, there are only a few role-oriented embedding approaches proposed recently. Although less explored, considering the importance of roles in analyzing networks and many applications that role-oriented NE can shed light on, it is necessary and timely to provide a comprehensive overview of existing role-oriented NE methods. In this review, we first clarify the differences between community-oriented and role-oriented network embedding. Afterwards, we propose a general framework for understanding role-oriented NE and a two-level categorization to better classify existing methods. Then, we select some representative methods according to the proposed categorization and briefly introduce them by discussing their motivation, development and differences. Moreover, we conduct comprehensive experiments to empirically evaluate these methods on a variety of role-related tasks including node classification and clustering (role discovery), top-k similarity search and visualization using some widely used synthetic and real-world datasets...

Multiple Social Role Embedding

Cross-Network Social User Embedding with Hybrid Differential Privacy Guarantees

Multi-Task Network Embedding

Embedding Identity and Interest for Social Networks.

Multi-Path Relationship Preserved Social Network Embedding.

Multi-Task Learning Based Network Embedding

Learning Multiple Network Embeddings for Social Influence Prediction

Disentangled Representation Learning for Structural Role Discovery

Role-based Network Embedding Via Structural Features Reconstruction with Degree-Regularized Constraint

Social Rank Regulated Large-scale Network Embedding

Multi-Task Network Representation Learning

Dynamic Structural Role Node Embedding for User Modeling in Evolving Networks

Network Embedding Via Coupled Kernelized Multi-Dimensional Array Factorization.

User Identification Cross Multiple Social Media Platform with Revised Input Output Network Embedding Framework

Modeling Large-Scale Dynamic Social Networks Via Node Embeddings

Multi-View Collaborative Network Embedding

A Multi-Semantic Metapath Model for Large Scale Heterogeneous Network Representation Learning

Multi-network User Identification via Graph-Aware Embedding.

Aligning Users Across Social Networks Using Network Embedding

A Survey on Role-Oriented Network Embedding

Network Embedding Based Recommendation Method in Social Networks.