Graph Neural Networks for Vehicular Social Networks: Trends, Challenges, and Opportunities
Elham Binshaflout, Student Member, IEEE , Aymen Hamrouni, Student Member, IEEE , and Hakim Ghazzai, Senior Member, IEEE · 2025
Read Graph Neural Networks for Vehicular Social Networks: Trends, Challenges, and Opportunities by Elham Binshaflout, Student Member, IEEE , Aymen Hamrouni, Student Member, IEEE , and Hakim Ghazzai, Senior Member, IEEE (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-14720.txt.
Introduction
Elham Binshaflout, Aymen Hamrouni, and Hakim Ghazzai survey how graph neural networks are being used in vehicular social networks, where cars, drivers, and road systems form a linked web. Networks of this kind suit graph methods well, since traffic, weather, and road users all interact. The authors say theirs is the first full review focused on this pairing.
The survey sorts prior studies by task, covering flow and path prediction, forecasting, signal control, driving help, routing, and connection management. It weighs both road geometry data and more abstract linked data. Gaps and open problems are flagged, giving new researchers a map of where vehicular network work is heading.
Catalog record
| CREATOR | Elham Binshaflout, Student Member, IEEE , Aymen Hamrouni, Student Member, IEEE , and Hakim Ghazzai, Senior Member, IEEE |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 114 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-14720
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Graph Neural Networks for Vehicular Social Networks: Trends, Challenges, and Opportunities Elham Binshaflout, Student Member, IEEE , Aymen Hamrouni, Student Member, IEEE , and Hakim Ghazzai, Senior Member, IEEE Abstract Graph Neural Networks (GNNs) have emerged as powerful tools for modeling complex, interconnected data, making them particularly well suited for a wide range of Intelligent Transportation System (ITS) applications. This survey presents the first comprehensive review dedicated specifically to the use of GNNs within Vehicular Social Networks (VSNs). By leveraging both Euclidean and non-Euclidean transportation-related data, including traffic patterns, road users, and weather conditions, GNNs offer promising solutions for analyzing and enhancing VSN applications. The survey systematically categorizes and analyzes existing studies according to major VSN-related tasks, including traffic flow and trajectory prediction, traffic forecasting, signal control, driving assistance, routing problem, and connectivity management. It further provides quantitative insights and synthesizes key takeaways derived from the literature review. Additionally, the survey examines the available datasets and outlines open research directions needed to advance GNN-based VSN applications. The findings indicate that, although GNNs demonstrate strong potential for improving the accuracy, robustness, and real-time performances of on task-specific or sub-VSN graphs, there remains a notable absence of studies that model a complete, standalone VSN encompassing all functional components. With the increasing availability of data and continued progress in graph learning, GNNs are expected to play a central role in enabling future large-scale and fully integrated VSN applications. I Introduction Developing an efficient Intelligent Transportation System (ITS) is becoming increasingly demanding due to several factors including, but not limited to, growing urbanization, increased vehicle ownership, and environmental concerns. The rapid increase in traffic volume and industrial city development has also made transitioning from traditional to smart cities increasingly urgent. This involves establishing integrated communication channels among road users and enabling simultaneous processing of tasks, including transportation management and control, seamless vehicle communication, and safety and privacy enhancement [ 68 ] . Within this framework, Vehicular Social Networks (VSNs) emerge as a specialized subset, distinctly enabling real-time communication between various road entities such as vehicles, road users, riders, and road units. VSNs integrate Social Networks (SNs) and Vehicular Ad-hoc Networks (VANETs) to establish a highly efficient platform for data dissemination. As a cutting-edge advancement within the Internet of Vehicles (IoV), VSNs provide a unique platform for reliable communication and collaborative decision-making among vehicles and road participants. This technology aims to enhance driving safety, efficiency, and comfort by enabling vehicles to share information about traffic conditions, road hazards, and driving behaviors in real time [ 101 ] . VSNs leverage these interactive capabilities to alleviate traffic congestion, reduce emissions, and improve the responsiveness of drivers, thus offering targeted solutions that complement the overarching goals of ITS [ 126 ] . The role of VSNs in driving forward these intelligent capabilities highlights their potential to shape the future of vehicular communication significantly. Moreover, the integration of VSNs within ITS is further justified by their capacity to handle and analyze the high volume of data generated through these interactive lines of communication. With the increasing availability of high-dimensional real-time traffic data, deep-learning-based approaches, particularly those utilizing graph neural networks, have been proposed to enhance VSN-related tasks [ 48 , 18 , 135 ] . These models are crucial for capturing and analyzing historical traffic data flows, understanding the relationships among traffic participants, and accurately representing the dynamic spatial-temporal metrics of traffic networks. Hence, there is a challenge to accurately represent the traffic features, road network structure, and road users’ relationships while extracting the topological relations and complex correlations among the connected entities. TABLE I: Existing surveys on GNNs for ITS and related fields and comparisons with the proposed work Title Ref. Year Main Focus Limitations A Survey on Vehicular Social Networks [ 101 ] 2015 Explores established trends in integrating connected vehicles with social networks. Absence of graph learning or GNN-based modeling for VSN applications. A Survey on Platoon-Based Vehicular Cyber-Physical Systems [ 41 ] 2016 Reviews platoon-based vehicular CPS, including techniques and communication standards. Limited focus on VSNs; lack of social interactions, graph-based modeling, or GNN methods. Vehicular Social Networks: A Survey [ 85 ] 2018 Reviews VSN architectures, communication models, and mobility patterns. Lack of graph-learning and absence of GNN-based analysis. A Survey on Pseudonym Changing Strategies for VANETs [ 5 ] 2018 Covers security and privacy-preserving strategies in VANETs. Lack of VSN context; absence of graph-learning and GNN-based approaches. Graph Neural Networks: A Review of Methods and Applications [ 134 ] 2020 Comprehensive overview of GNN architectures and applications across domains. Does not address ITS or VSN applications; no coverage of transportation datasets. Network Representation Learning: A Macro and Micro View [ 60 ] 2021 Analyzes network embedding and graph-based representation learning methods. Absence of ITS/VSN applications; does not explore GNN-based solutions. Machine Learning for Next-Generation Intelligent Transportation Systems [ 99 ] 2021 Reviews ML techniques in ITS (cooperative driving, hazard detection, etc.). Lack of graph-learning perspective; does not address GNN or VSN-specific modeling. Graph Neural Networks: Architectures, Stability, and Transferability [ 90 ] 2021 Focuses on GNN robustness and transferability across graph structures. Does not explore ITS or VSN tasks; no discussion of mobility or social graphs. A Comprehensive Survey on Graph Neural Networks [ 113 ] 2021 Compares GNN architectures, datasets, open-source implementations, and applications. Does not address ITS or VSN domains. Graph Neural Networks in Recommender Systems: A Survey [ 110 ] 2022 Reviews GNN-based recommender systems and their learning paradigms. No coverage of transportation or VSN applications. Graph Neural Network for Traffic Forecasting: A Survey [ 43 ] 2022 Summarizes spatio-temporal GNN models for traffic forecasting. Limited to forecasting; lack of VSN context; absence of social-interaction graph modeling. Data-Centric Approaches in the Internet of Vehicles [ 80 ] 2023 Reviews data management and analysis in IoV. Focus on data management; lack of GNN or social-vehicular graph modeling. The Internet of Vehicles and Sustainability: Reflection on Environmental, Social, and Corporate Governance [ 49 ] 2023 Proposing a sustainable model for IoV design, mapping Environmental, Socia,l and Governance factors Absence of graph learning or GNN modeling Graph Neural Networks for Traffic Pattern Recognition [ 4 ] 2023 Overview of GNNs for traffic pattern analysis. Limited to one domain; lack of VSN-level graph modeling or multi-domain coverage. Graph Neural Networks for Time Series [ 45 ] 2024 Analyzes GNN architectures for time-series forecasting, classification, and anomaly detection. Does not address ITS or VSN tasks; absence of social or mobility graph modeling. Graph Neural Networks in Intelligent Transportation Systems [ 51 ] 2024 Reviews GNN applications across six ITS domains. Lack of focus on VSN; limited social-interaction graph modeling; datasets not categorized by VSN-specific tasks. Graph Neural Networks for Vehicular Social Networks: Trends, Challenges and Opportunities Proposed - Comprehensive review and analysis of VSNs and the GNNs architectures. Present an overview of promising GNN-based VSN applications, structures, and datasets. Bridges the gap between VSN and GNN research by jointly discussing both domains and providing a comprehensive review of GNN-based VSN applications, categorized datasets, and key future opportunities. With the high volume and heterogeneity of the data, this task becomes non-trivial to handle by vanilla deep learning technologies. As a matter of fact, such traditional