About the Special Issue

Graph data has emerged as the backbone of intelligent decision-making systems across numerous domains. Recent studies indicate that graph-based approaches have seen a surge increase in implementation across AI systems in the past five years, highlighting their growing significance in the field.

This special issue explores how emerging technologies (ET) are revolutionizing traditional graph data processing and analysis methods (PAGD). These innovations provide not only new theoretical frameworks but also practical solutions for tackling large-scale, dynamic, and noisy complex graph data challenges. Research in ET-PAGD offers both significant theoretical value and extensive practical applications, creating a robust foundation for digital transformation and intelligent upgrading across industries.

The Landscape of Graph Data Processing

PAGD encompasses diverse graph types, each with unique structural characteristics:

• Static Graphs: Feature stable topologies, driving research in structure-based querying, mining, and reasoning algorithms
• Dynamic Graphs: Evolve through node/edge additions or removals, enabling research in incremental computation and temporal modeling for real-time analysis
• Hypergraphs: Model complex multidirectional associations, offering new perspectives for representing higher-order interactions
• Noisy Graphs: Contain redundant or erroneous information, necessitating techniques to enhance data quality and algorithmic robustness

Research Challenges and Opportunities

Despite significant advancements in PAGD technologies, substantial challenges remain when handling massive, complex, and dynamically changing graph data:

• Improving algorithmic robustness while maintaining computational efficiency
• Designing adaptive processing strategies for different graph data types
• Effectively integrating deep learning and reinforcement learning methods into PAGD
• Developing scalable solutions for industry-scale graph problems

These challenges present exciting research opportunities to develop reliable, robust, and efficient ET-PAGD approaches. To share the most recent advances, current challenges and potential applications of theories and methods for ET-PAGD, we are delighted and honored to propose this special issue of Artificial Intelligence Research.

Topics of Interest

This special issue welcomes high-quality submissions on emerging techniques for processing and analyzing graph data, including but not limited to:

• Graph Matching and Similarity Computation
• Graph Partitioning and Decomposition
• Graph Clustering and Community Detection
• Graph Reduction and Compression
• Graph Embedding and Representation Learning
• Graph Neural Networks
• Graph Structure Learning
• Self-supervised Learning on Graphs
• Graph Sampling Methodologies
• Graph Generation and Synthesis
• Graph Denoising and Cleaning
• Link Prediction and Completion
• Social Network Analysis
• Biomedical Graph Analytics
• Financial Risk Control
• Network Security
• Urban Planning and Smart Cities
• Knowledge Graphs and Reasoning
• Other theories, methods, algorithms, and applications related to ET-PAGD;

Important Dates

• Expected first submission: December 31, 2025
• First review round completed: February 28, 2026
• Revised manuscripts due: March 31, 2026
• Completion of the review and revision process (final notification): May 31, 2026

Guest Editors

• Yulin He (yulinhe@gml.ac.cn), Ph.D.
  Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ), Shenzhen 518107, China
  ORCID: 0000-0002-3415-0686
  https://scholar.google.ca/citations?user=8rsSv7oAAAAJ&hl=en&oi=ao 

• Jiaxing Shen (jiaxingshen@ln.edu.hk), Ph.D.
  School of Data Science, Lingnan University, Hong Kong, China
  ORCID: 0000-0002-0833-0288
  https://scholar.google.com/citations?user=amJMlVoAAAAJ&hl=en

• Joshua Zhexue Huang (huang@szu.edu.cn), Ph.D.
  Big Data Institute, College of Computer Science and Software Engineering, Shenzhen University, Shenzhen 518107, China