Big Data Mining & Intelligent Computing

Overview

We research efficient graph stream algorithms, game-theoretic optimization in social networks, personalized recommendation systems, and multimodal knowledge graph completion for large-scale network data and multi-source heterogeneous information, providing theoretical and methodological support for data-driven scientific discovery.

Research Directions

• Graph Stream Computing: Distributed triangle counting and subgraph matching algorithms in streaming environments
• Social Network Optimization: Game-theoretic social welfare control and incentive mechanisms in vehicular crowdsensing
• Personalized Recommendation: Trip recommendation and route analysis based on graph attention auto-encoders
• Knowledge Graph Completion: Knowledge representation-enhanced multimodal knowledge graph completion methods
• Data Fabric Storage: Key technologies for data fabric distributed storage

Key Achievements

• Proposed distributed streaming triangle counting framework (KBS 2023, ACM TKDD 2021)
• Proposed compact estimator for streaming triangle counting (IEEE TKDE 2024)
• Proposed unilateral control for social welfare in vehicular crowdsensing iterated games (JCST 2025)
• Proposed graph attention auto-encoder for personalized trip recommendation (KBS 2025, AAAI 2020)
• Proposed knowledge representation-enhanced multimodal knowledge graph completion (Computer Engineering and Applications 2025)
• Proposed social welfare control and free-riding incentives in cross-silo federated learning caching (ICASSP 2026)