Work

Research on cooperative relationship prediction based on transformer dynamic graph modeling and comparative learning

Year: 2025

Type: article

Abstract: The goal of this study is to develop an accurate and efficient framework for predicting scientific research cooperation relationships in dynamic academic networks. Existing temporal graph learning models often struggle with sparse and heterogeneous collaboration data, limiting their applicability in real-world academic settings. To address this challenge, we propose the transformer-based contrastive learning (TCL) model, which integrates a topology-aware transformer with a dual-stream encoder to capture structural and temporal dependencies in evolving collaboration graphs jointly. We construct dynamic network snapshots from two large-scale datasets: National Key R&D Project (NKPD) collaboration networks and citation network data (CND). Comparative experiments against state-of-the-art baselines, including DySAT, TGAT, TGN, and CLDG, demonstrate that TCL achieves the highest accuracy-up to 95.3%area under the ROC curve (AUC)—while reducing training time by 21%. Furthermore, our analysis reveals that network sparsity and time-slice granularity significantly influence prediction performance, with annual slices outperforming half-year slices by 35.8% in AUC. The impact of this work lies in providing a robust and scalable tool for forecasting research partnerships, enabling funding agencies, institutions, and policymakers to identify high-potential collaborations earlier. This contributes to accelerating interdisciplinary innovation and enhancing the strategic allocation of research resources. • We use the Transformer dynamic diagram modeling based on control learning model (TCL) to empirically investigate the prediction of scientific research partnerships based on national key project paper collaboration data and citation network data. • The empirical experiments show that the choice of time intervals and network density have an important impact on the accuracy of the prediction results of the dynamic research network, and the TCL model shows high accuracy and efficiency in dynamic link prediction with the year as the time interval.

Source: International Journal of Cognitive Computing in Engineering

DOI: 10.1016/j.ijcce.2025.10.002

Authors: Wei Yu, Ao Sun, Hongyu Wang, Dandan Chen, Long Zhang, Yanxia Zhao

Institutions: Zhejiang Yuexiu University, Zhejiang University of Science and Technology, Tianjin University

Language: English

Topic: Advanced Graph Neural Networks

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: gold

Estimated APC paid: $700