Influence Maximization in Multi-layer Social Networks Based on Differentiated Graph Embeddings

2025-08-13

AI Summary

Existing methods for identifying critical nodes in multiplex social networks suffer from three key limitations: (1) neglect of local opinion leaders, causing seed node overlap; (2) inability of standard GNNs to effectively aggregate heterogeneous influence strengths across layers; and (3) failure to jointly model multiplex structural properties and node heterogeneity. To address these issues, we propose a multiplex graph neural network framework tailored for influence maximization. Its core contributions are: (1) self-influence propagation subgraph construction, explicitly capturing inter-layer heterogeneity and intra-layer diffusion dynamics; (2) adaptive local influence aggregation, mitigating seed overlap while enhancing fusion of heterogeneous features; and (3) differential graph embedding design, aligning node representations with underlying propagation dynamics. Extensive experiments on four real-world multiplex social network datasets demonstrate that our method significantly outperforms state-of-the-art influence maximization algorithms, achieving substantial improvements in cascade size prediction accuracy.

Technology Category

• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Search and Optimization: Local Search
• Machine Learning: Graph-based Machine Learning

Application Category

• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Identifying influential nodes is crucial in social network analysis. Existing methods often neglect local opinion leader tendencies, resulting in overlapping influence ranges for seed nodes. Furthermore, approaches based on vanilla graph neural networks (GNNs) struggle to effectively aggregate influence characteristics during message passing, particularly with varying influence intensities. Current techniques also fail to adequately address the multi-layer nature of social networks and node heterogeneity. To address these issues, this paper proposes Inf-MDE, a novel multi-layer influence maximization method leveraging differentiated graph embedding. Inf-MDE models social relationships using a multi-layer network structure. The model extracts a self-influence propagation subgraph to eliminate the representation bias between node embeddings and propagation dynamics. Additionally, Inf-MDE incorporates an adaptive local influence aggregation mechanism within its GNN design. This mechanism dynamically adjusts influence feature aggregation during message passing based on local context and influence intensity, enabling it to effectively capture both inter-layer propagation heterogeneity and intra-layer diffusion dynamics. Extensive experiments across four distinct multi-layer social network datasets demonstrate that Inf-MDE significantly outperforms state-of-the-art methods.

Problem

• Identifies influential nodes in multi-layer social networks
• Addresses overlapping influence ranges of seed nodes
• Handles varying influence intensities and node heterogeneity

Innovation

• Multi-layer network structure modeling
• Self-influence propagation subgraph extraction
• Adaptive local influence aggregation mechanism