A Method for Predicting Links in Complex Networks by Integrating Enclosure Subgraphs With High-Frequency Graph Information
Open Access
- 2 May 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 63209-63222
Abstract
Link prediction in complex networks, crucial for uncovering hidden or upcoming links between nodes and widely applicable in fields such knowledge graphs, faces challenges with current techniques. Predominantly, graph neural networks (GNN) based methods focus on learning node representations and use predictive components to assess the similarity of these representations for achieving link prediction. However, these approaches tend to accumulate errors in the predictive model and complicates the training process. Additionally, existing GNNs often display a low-pass filtering effect during network data processing, prioritizing low-frequency information while overlooking high-frequency details in node representations. These bias make GNNs mainly used for link prediction in strongly assortative networks and limit their performance on highly disassortative networks. Addressing these issues, this article introduces a novel framework that redefines the link prediction problem. By extracting enclosure subgraphs of both ‘observed’ and ‘unobserved’ links, we represent these links by corresponding enclosure subgraphs and transform link prediction into a problem of subgraphs classification. We innovate by combining high- and low-frequency graph information from the subgraphs, using an attention mechanism for integration, and constructing a graph neural network tailored to learn these subgraph representations, thus accomplishing the task of link prediction indirectly and enhancing link subgraphs classification accuracy. Our extensive experiments on recognized benchmark datasets, evaluated using the $Hits\text{@}n$ metric, demonstrate that our method not only shows remarkable performance but also possesses strong generalization capabilities, positioning it as a potent baseline for link prediction tasks.Keywords
Funding Information
- Natural Science Foundation of Anhui Province (1908085QF283)
- Doctoral Scientific Research Funding (2021bsk016, 2023bsk024)
- University Synergy Innovation Program of Anhui Province (GXXT-2022-047)
- Anhui Provincial Universities’ Excellent Young Teachers Training Program (YQYB2023053)
- Natural Science Research Projects in Universities (2023AH040314)
- Suzhou University Scientific Research Funding (2021XJPT50)
This publication has 21 references indexed in Scilit:
- A Graph Attention Network-Based Link Prediction Method Using Link Value EstimationIEEE Access, 2023
- Exploring an edge convolution and normalization based approach for link prediction in complex networksJournal of Network and Computer Applications, 2021
- STRING v11: protein–protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasetsNucleic Acids Research, 2018
- Graph embedding techniques, applications, and performance: A surveyKnowledge-Based Systems, 2018
- Network Representation Learning: A SurveyIEEE Transactions on Big Data, 2018
- DrugBank 5.0: a major update to the DrugBank database for 2018Nucleic Acids Research, 2017
- Mining overlapping and hierarchical communities in complex networksPhysica A: Statistical Mechanics and its Applications, 2015
- Detecting overlapping communities in networks using the maximal sub-graph and the clustering coefficientPhysica A: Statistical Mechanics and its Applications, 2014
- Two-stage algorithm using influence coefficient for detecting the hierarchical, non-overlapping and overlapping community structurePhysica A: Statistical Mechanics and its Applications, 2014
- A new model for learning in graph domains2006