contextual subgraph extraction

This work addresses the limited expressiveness of conventional temporal triplet queries in temporal networks by proposing a first-order logic (FOL)-based triplet query framework supporting thresholded quantification. The framework uniquely unifies existential and thresholded universal quantifiers within temporal graph querying, significantly enhancing the semantic expressiveness of motif discovery. To realize this, the authors introduce FOLTY—a novel, efficient algorithm that integrates temporal-aware indexing, sparse-graph-optimal traversal strategies, and quantified condition verification mechanisms, achieving theoretically optimal time complexity. Experimental evaluation demonstrates that FOLTY processes temporal graphs with up to 70 million edges within one hour on commodity hardware, exhibiting superior performance and strong scalability.

To address the parameter sensitivity, high global traversal overhead, and inefficiency in online retrieval inherent in hypergraph cohesive subgraph discovery, this paper proposes the first locality-aware indexing framework for hypergraph cohesive subgraphs. Departing from conventional reliance on preset parameters and full-graph traversal, our framework encodes local hypergraph structures, constructs multi-granularity indexes, and employs a neighborhood-propagation-driven pruning query algorithm—enabling sublinear-time, diverse cohesive subgraph retrieval. Evaluated on multiple real-world datasets, our method achieves an average 12.6× speedup and 43% memory reduction over state-of-the-art baselines, while supporting millisecond-level dynamic queries. This significantly enhances both the practicality and scalability of higher-order relational modeling.

To address inefficient graph data extraction and inaccurate intent matching from relational databases, this paper proposes ExtGraph—a user-intent-driven, efficient graph extraction method. ExtGraph innovatively integrates outer joins with materialized views to establish a hybrid query processing mechanism that precisely reconstructs user-specified graph structures under complex multi-table join scenarios. It employs query rewriting, precomputed materialized views, and lightweight optimization strategies to substantially reduce graph construction overhead. Experimental evaluation on TPC-DS, DBLP, and IMDb datasets demonstrates that ExtGraph achieves up to 2.78× speedup over state-of-the-art approaches, while preserving both structural integrity and semantic fidelity of the extracted graphs. This work establishes a novel, efficient, and scalable paradigm for end-to-end relational-to-graph transformation, bridging the gap between relational data management and graph analytics.

Subgraph counting on graph data faces an inherent trade-off between expressive power and computational efficiency. Method: This paper proposes the Localized Weisfeiler–Leman (Local k-WL) framework, introducing a novel subgraph fragmentation decomposition technique that enables exact counting of all induced subgraphs of size ≤4 using only 1-WL. The approach integrates a three-tier differentiable learning architecture, bridging combinatorial algorithms with end-to-end GNN training, and rigorously proves its expressive power lies strictly between k-WL and (k+1)-WL. Contribution/Results: Compared to standard k-WL, Local k-WL achieves significantly lower time and space complexity. Experiments on computational biology and social network datasets demonstrate superior performance in counting accuracy, generalization, and inference efficiency—establishing a new state-of-the-art for scalable, expressive subgraph counting.

Existing graph databases lack effective support for the tree-shaped substructures commonly found in property graphs. This work addresses this limitation by treating such tree substructures as first-class citizens and proposes a systematic management framework encompassing modeling, indexing, and query optimization. Drawing inspiration from XML structural indexing techniques, the approach enables efficient path queries within a relational graph database backend. Experimental evaluation demonstrates that the proposed method significantly improves path query performance, thereby validating the potential of structural indexing to enhance graph data management.

This work investigates how to achieve approximately uniform edge sampling in sublinear time under a hybrid query model that combines independent set queries and local graph queries. It establishes, for the first time, a tight two-way reduction between edge sampling and approximate edge counting in this model, demonstrating that the two problems share matching upper and lower bounds in query complexity. The proposed sampling algorithm matches the query complexity of the current best-known edge counting algorithms, while the analysis yields tight lower bounds for each constituent query model. These results highlight the pivotal role of independent set queries in enabling efficient graph sampling tasks, revealing a fundamental equivalence between sampling and counting in terms of computational hardness under the hybrid query framework.

Existing graph analysis systems struggle to effectively integrate topological structure with node attributes, limiting the discovery of patterns driven by their interaction. This work proposes ZipLine, a novel system that, for the first time, unifies predicate logic to express topology, node attributes, and neighborhood relationships within a single formalism. ZipLine introduces an interaction-driven predicate learning algorithm that enables cross-space collaborative reasoning and iterative analysis. By integrating coordinated views, subgraph selection, and attribute brushing techniques, the system facilitates expressive and efficient exploration of complex patterns in multivariate graphs. Empirical evaluation across three real-world domains—energy infrastructure, cybersecurity, and drug discovery—demonstrates ZipLine’s effectiveness in significantly enhancing the expressiveness and discoverability of intricate graph patterns.

This work addresses the problem of efficiently detecting an arbitrarily structured subgraph embedded in a random graph under the constraint of only a limited number of non-adaptive edge queries. By establishing matching information-theoretic lower bounds and algorithmic upper bounds, the paper presents the first unified framework for query complexity applicable to general embedded subgraphs, extending classical full-observation models to information-constrained settings. Leveraging structural properties such as local dense motifs, high-degree vertices, and global edge density, the authors design detection algorithms whose query complexities are tight up to polylogarithmic factors across broad classes of subgraphs—including clique-like, bounded-cover, and hub-dominated structures—thereby achieving near-optimal performance in terms of query efficiency.

Subgraph extraction problems arise widely in network design, facility location, and related domains, yet lack a general-purpose, efficient solution methodology. This work proposes ΔSearch—the first unified heuristic framework that requires only user-specified feasibility constraints and an optimization objective, automatically adapting to monotone, weighted monotone, and non-monotone graph problems without problem-specific parameter tuning. By integrating a reward-penalty optimization mechanism, generic constraint modeling, and search space pruning techniques, ΔSearch substantially enhances computational efficiency and can accelerate exact algorithms. Empirical evaluations demonstrate that it matches or surpasses state-of-the-art heuristics on tasks such as maximum planar subgraph, uncapacitated facility location, and prize-collecting vertex cover, while achieving approximately 89% of optimal solution quality on average across other problems—all without any parameter tuning.

This study addresses a central question in network science: how ubiquitous global structural features of complex networks—such as hubs, short path lengths, and high clustering—emerge without access to global information. The work proposes that these macroscopic properties arise not from global mechanisms but through bottom-up emergence driven by simple local rules, wherein nodes connect solely based on information from their immediate neighbors. By constructing a purely local growth model, analyzing empirical networks across diverse domains—including citation, social, and protein–protein interaction networks—and providing intuitive theoretical explanations, the study demonstrates for the first time that a unified local rule can reproduce key topological characteristics of real-world networks without invoking global assumptions such as preferential attachment. This finding offers a new paradigm for understanding self-organization in complex systems.