Predicting co-word links via heterogeneous graph convolutional networks

Abstract Co-word analysis, which explores the co-occurrence of key terminology within a specific field, is a valuable tool for identifying research themes and their networks. Leveraging the booming machine learning models, link prediction in co-word networks makes it possible to discover potential i...

Full description

Saved in:
Bibliographic Details
Main Authors: Yangmin Li, Xin Zhang, Xin Bai, Sen Bai, Zhengang Jiang
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-05853-w
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849334783353552896
author Yangmin Li
Xin Zhang
Xin Bai
Sen Bai
Zhengang Jiang
author_facet Yangmin Li
Xin Zhang
Xin Bai
Sen Bai
Zhengang Jiang
author_sort Yangmin Li
collection DOAJ
description Abstract Co-word analysis, which explores the co-occurrence of key terminology within a specific field, is a valuable tool for identifying research themes and their networks. Leveraging the booming machine learning models, link prediction in co-word networks makes it possible to discover potential interactions between research themes and reveal emerging trends. Nevertheless, few existing methods have explored end-to-end deep models, impeded by the limitations of text graph models in learning both word co-occurrence and word-document relations implicit in co-word networks simultaneously. In this work, we propose to use a heterogeneous graph convolutional network (GCN) modeling to jointly learn word embeddings and document embeddings directly from co-word networks, incorporating document-specific information. The learning model is supervised by the binary labels for the existence of co-word links. Extensive experiments have been conducted on the Web of Science dataset from Information Science and Library Science. Experimental results show that the AUC value of our GCN-based approach is $$93.46\%$$ , whereas the AUC value of the best traditional machine learning method is $$89.15\%$$ .
format Article
id doaj-art-91f3d239bc594cd6ab09f316164e2e6c
institution Kabale University
issn 2045-2322
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-91f3d239bc594cd6ab09f316164e2e6c2025-08-20T03:45:28ZengNature PortfolioScientific Reports2045-23222025-07-0115111410.1038/s41598-025-05853-wPredicting co-word links via heterogeneous graph convolutional networksYangmin Li0Xin Zhang1Xin Bai2Sen Bai3Zhengang Jiang4School of Computer Science and Technology, Changchun University of Science and TechnologySchool of Computer Science and Technology, Changchun University of Science and TechnologyResearch Department, Huawei Technologies Co. LtdSchool of Computer Science and Technology, Changchun University of Science and TechnologySchool of Computer Science and Technology, Changchun University of Science and TechnologyAbstract Co-word analysis, which explores the co-occurrence of key terminology within a specific field, is a valuable tool for identifying research themes and their networks. Leveraging the booming machine learning models, link prediction in co-word networks makes it possible to discover potential interactions between research themes and reveal emerging trends. Nevertheless, few existing methods have explored end-to-end deep models, impeded by the limitations of text graph models in learning both word co-occurrence and word-document relations implicit in co-word networks simultaneously. In this work, we propose to use a heterogeneous graph convolutional network (GCN) modeling to jointly learn word embeddings and document embeddings directly from co-word networks, incorporating document-specific information. The learning model is supervised by the binary labels for the existence of co-word links. Extensive experiments have been conducted on the Web of Science dataset from Information Science and Library Science. Experimental results show that the AUC value of our GCN-based approach is $$93.46\%$$ , whereas the AUC value of the best traditional machine learning method is $$89.15\%$$ .https://doi.org/10.1038/s41598-025-05853-wCo-word networksGraph convolutional networksLink prediction
spellingShingle Yangmin Li
Xin Zhang
Xin Bai
Sen Bai
Zhengang Jiang
Predicting co-word links via heterogeneous graph convolutional networks
Scientific Reports
Co-word networks
Graph convolutional networks
Link prediction
title Predicting co-word links via heterogeneous graph convolutional networks
title_full Predicting co-word links via heterogeneous graph convolutional networks
title_fullStr Predicting co-word links via heterogeneous graph convolutional networks
title_full_unstemmed Predicting co-word links via heterogeneous graph convolutional networks
title_short Predicting co-word links via heterogeneous graph convolutional networks
title_sort predicting co word links via heterogeneous graph convolutional networks
topic Co-word networks
Graph convolutional networks
Link prediction
url https://doi.org/10.1038/s41598-025-05853-w
work_keys_str_mv AT yangminli predictingcowordlinksviaheterogeneousgraphconvolutionalnetworks
AT xinzhang predictingcowordlinksviaheterogeneousgraphconvolutionalnetworks
AT xinbai predictingcowordlinksviaheterogeneousgraphconvolutionalnetworks
AT senbai predictingcowordlinksviaheterogeneousgraphconvolutionalnetworks
AT zhengangjiang predictingcowordlinksviaheterogeneousgraphconvolutionalnetworks