N-ary Knowledge Graphs (NKGs) are a specialized type of knowledge graph designed to efficiently represent complex real-world facts.Unlike traditional knowledge graphs, where a fact typically involves two entities, NKGs can capture n-ary facts containing more than two entities.Link prediction in NKGs aims to predict missing elements within these n-ary facts, which is essential for completing NKGs and improving the performance of downstream applications.This task has recently gained significant attention.In this paper, we present the first comprehensive survey of link prediction in NKGs, providing an overview of the field, systematically categorizing existing methods, and analyzing their performance and application scenarios.We also outline promising directions for future research.

In many real-world applications, the evolving relationships between entities can be modeled as temporal graphs, where each edge has a timestamp representing the interaction time. As a fundamental problem in graph analysis, community search (CS) in temporal graphs has received growing attention but exhibits two major limitations: (1) Traditional methods typically require predefined subgraph structures, which are not always known in advance. (2) Learning-based methods struggle to capture temporal interaction information. To fill this research gap, in this paper, we propose an effective Unsupervised Temporal Community Search with pre-training of temporal dynamics and subgraph knowledge model (UTCS ). UTCS contains two key stages: offline pre-training and online search. In the first stage, we introduce multiple learning objectives to facilitate the pre-training process in the unsupervised learning setting. In the second stage, we identify a candidate subgraph and compute community scores using the pre-trained node representations and a novel scoring mechanism to determine the final community members. Experiments on five real-world datasets demonstrate the effectiveness of the proposed method.

Zixuan Li, Yutao Zeng, Yuxin Zuo, Weicheng Ren, Wenxuan Liu, Miao Su, Yucan Guo, Yantao Liu, Xiang Li, Zhilei Hu, Long Bai, Wei Li, Yidan Liu, Pan Yang, Xiaolong Jin, Jiafeng Guo, Xueqi Cheng. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Core decomposition is a critical metric for evaluating the vertex importance and analyzing graph structure. Given a graph$G$, a k-core is the largest subgraph of$G$where each vertex has at least$k$neighbors. Most existing works mainly focus on homogeneous graphs in which edges are of the same type and cannot be applied to heterogeneous information networks (HINs) directly. However, most real-world networks are HINs which consist of different vertex types and edge types. To reveal the cohesive subgraphs with hierarchical relations on HINs, we adopt the well-known$(k,\mathcal{P})$-core model to compute coreness over HINs, where$\mathcal{P}$is a meta-path, i.e., a sequence of relations defined between different types of vertices. Hence, the$(k,\mathcal{P})$-core is a subgraph where each vertex is connected to at least$k$other vertices via instances of$\mathcal{P}$. Based on two kinds of sparse matrix products, we propose two kinds of algebraic core decomposition algorithms, which are suitable for general HINs and locally dense HINs, respectively. We have performed extensive empirical evaluations of our algorithms on six large real-world HINs. The results show that the proposed solutions are highly efficient for core decomposition and achieve up to$258.84\times$speedup than the state-of-the-art parallel algorithm on 20 cores. Moreover, other HIN tasks that involve homogeneous graph construction can also benefit from our algorithms.