Graph pattern matching is a fundamental task in various fields, enabling the exploration of complex graph structures. Existing graph pattern matching systems focus on generating better schedule plan to filter out invalid matching paths through pattern analysis. However, these systems ignore the numerous identical computations among different filtered matching paths.To overcome this challenge, we propose Reduced Redundancy Computation (R2Mine), aims to effectively recognize and reuse identical computations among different filtered matching paths. Specifically, by analyzing pattern, R2Mine extracts overlap nodes among all valid matching instances, named Reusable Basic Pattern (RBP). Then R2Mine employs RBP to generate redundant discriminant conditions to recognize identical computations involved overlap nodes. Utilizing these conditions allows the system to reuse redundant computations among different matching paths. To evaluate our approach, we conduct extensive experiments on 6 real-world graph datasets. The results demonstrate that R2Mine outperforms state-of-the-art graph pattern matching systems, Peregrine and GraphPi, respectively achieving a 13.61× and 1.75× on average improvement in performance.

As data scales continue to expand, graph structures are widely applied across multiple domains due to their effective organization of complex data. Graph Pattern Matching (GPM) is a fundamental task in graph analysis to identify all user-interesting subgraphs in a graph. Current GPM systems achieve this goal by generating efficient traversal path strategies. However, when matching patterns that include a specific vertex (S-GPM), current GPM systems often traverse paths without the specific vertex or duplicate traverse some paths. These redundant traversals lead to decreased execution efficiency. In this paper, we introduce AnchorMine, a GPM system designed for S-GPM tasks, aiming to significantly reduce redundant path traversal by identifying and reusing paths that include specific vertex. Specifically, AnchorMine first analyzes the pattern to identify vertices in different positions within the pattern, named Anchors (ACs). Then AnchorMine extracts features of reusable paths based on each Anchor (AC). These features enable the system to identify paths that can be reused during matching. Using these features, it further generates the parameters required for matching based on path reuse, achieving efficient matching for S-GPM tasks. In experiments on 8 real-world graph datasets, AnchorMine significantly outperformed GraphPi, SandSlash, and Peregrine on 6 datasets used for performance testing, with matching performance improvements of 3249.22 ×, 2018.73 × and 7573.27 ×, respectively. On the remaining 2 datasets used for scalability testing, AnchorMine scales well.

Recommender systems play a significant role in online services and have attracted wide attention from both academia and industry. In this paper, we focus on an important, practical, but often overlooked task: domain-level zero-shot recommendation (DZSR). The challenge of DZSR mainly lies in the absence of collaborative behaviors in the target domain, which may be caused by various reasons, such as the domain being newly launched without existing user-item interactions, or users' behaviors being too sensitive to collect for training. To address this challenge, we propose a Transferable Interest Graph Embedding technique for Recommendations (Tiger). The key idea is to connect isolated collaborative filtering datasets with a knowledge graph tailored to recommendations, then propagate collaborative signals from public domains to the zero-shot target domain. The backbone of Tiger is the transferable interest extractor, which is a simple yet effective graph convolutional network (GCN) aggregating multiple hops of neighbors on a shared interest graph. We find that the bottom layers of GCN preserve more domain-specific information while the upper layers represent universal interest better. Thus, in Tiger, we discard the bottom layers of GCN to reconstruct user interest so that collaborative signals can be successfully propagated to other domains, and retain the bottom layers of GCN to include domain-specific information for items. Extensive experiments with four public datasets demonstrate that Tiger can effectively make recommendations for a zero-shot domain and outperform several alternative baselines.

Knowledge graph (KG) entity typing focuses on inferring possible entity type instances, which is a significant subtask of knowledge graph completion (KGC). Existing entity typing methods usually exploit the entity representation to model the transmission between entities and their types, which cannot fully explore the fine-grained entity typing on identifying the semantic type of an entity. To address these issues, we propose Neighborhood-Attention Neural Fine-Grained Entity Typing (AttEt), which considers the neighborhood information of the entities from KGs to bridge entities and their types together. In this paper, AttEt first develops a type-specific attention mechanism to aggregate the neighborhood knowledge of the given entity with type-specific weights. These weights are beneficial to capture various characteristics for different types of the entity, and further imply the complex correlation among these fine-grained types. Then, AttEt adaptively integrates the aggregated neighbor-level representation with entity inherent embedding to calculate the matching score between the entity and its candidate type. Besides, many entities are sparse in their relations with other entities in KGs, which makes the entity typing task more challenging. To solve this problem, we present a smooth strategy on relation-sparsity entities to improve the robustness of the model. Extensive experiments on two real-world datasets (Freebase and YAGO) show that AttEt significantly outperforms state-of-the-art baselines in the [email protected] by 2.11% on Freebase and by 8.42% on YAGO, respectively.

The core objective of modelling recommender systems from implicit feedback is to maximize the positive sample score sp and minimize the negative sample score sn, which can usually be summarized into two paradigms: the pointwise and the pairwise. The pointwise approaches fit each sample with its label individually, which is flexible in weighting and sampling on instance-level but ignores the inherent ranking property. By qualitatively minimizing the relative score sn − sp, the pairwise approaches capture the ranking of samples naturally but suffer from training efficiency. Additionally, both approaches are hard to explicitly provide a personalized decision boundary to determine if users are interested in items unseen. To address those issues, we innovatively introduce an auxiliary score bu for each user to represent the User Interest Boundary(UIB) and individually penalize samples that cross the boundary with pairwise paradigms, i.e., the positive samples whose score is lower than bu and the negative samples whose score is higher than bu. In this way, our approach successfully achieves a hybrid loss of the pointwise and the pairwise to combine the advantages of both. Analytically, we show that our approach can provide a personalized decision boundary and significantly improve the training efficiency without any special sampling strategy. Extensive results show that our approach achieves significant improvements on not only the classical pointwise or pairwise models but also state-of-the-art models with complex loss function and complicated feature encoding.