Dynamic Subgraph Matching via Cost-Model-based Vertex Dominance Embeddings (Technical Report)

2024-07-23

arXiv.org

AI Summary

This paper addresses the Continuous Subgraph Matching (CSM) problem over dynamic graph streams, aiming to support efficient incremental matching and real-time queries under frequent edge updates. To overcome the limitations of existing approaches—namely weak pruning capability and high maintenance overhead—we propose the Vertex-Dominance Embedding (VDE) framework and the Degree-Aware Star-structured Summary (DAS³). VDE is the first to model vertex-dominance relationships as compact embeddings, while DAS³ integrates degree-based grouping and range-aware pruning to significantly enhance higher-order vertex pruning. Our incremental graph maintenance algorithm enables localized updates, eliminating costly global recomputation. Extensive experiments on real-world and synthetic datasets demonstrate that our method reduces matching latency by 62%–89% and decreases maintenance overhead by 53%–77% compared to state-of-the-art baselines, confirming its efficiency, scalability, and practicality.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Data Stream Mining
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

In many real-world applications such as social network analysis, knowledge graph discovery, biological network analytics, and so on, graph data management has become increasingly important and has drawn much attention from the database community. While many graphs (e.g., Twitter, Wikipedia, etc.) are usually involving over time, it is of great importance to study the dynamic subgraph matching (DSM) problem, a fundamental yet challenging graph operator, which continuously monitors subgraph matching results over dynamic graphs with a stream of edge updates. To efficiently tackle the DSM problem, we carefully design a novel vertex dominance embedding approach, which effectively encodes vertex labels that can be incrementally maintained upon graph updates. Inspire by low pruning power for high-degree vertices, we propose a new degree grouping technique over basic subgraph patterns in different degree groups (i.e., groups of star substructures), and devise degree-aware star substructure synopses (DAS^3) to effectively facilitate our designed vertex dominance and range pruning strategies. We develop efficient algorithms to incrementally maintain dynamic graphs and answer DSM queries. Through extensive experiments, we confirm the efficiency of our proposed approaches over both real and synthetic graphs.

Problem

• Efficiently monitors subgraph matches in dynamic graphs
• Handles continuous updates to graph edge streams
• Improves pruning for high-degree vertices via embeddings

Innovation

• Dynamic Vertex Dominance Embeddings for subgraph matching
• Degree grouping technique for star pattern decomposition
• Degree-aware star substructure synopses with pruning strategies