TBDRI: block decomposition based on relational interaction for temporal knowledge graph completion

Yu, Mo; Guo, Jiujiang; Yu, Jian; Xu, Tianyi; Zhao, Mankun; Liu, Hongwei; Li, Xuewei; Yu, Ruiguo

doi:10.1007/s10489-022-03601-5

Cited by 9 publications

(7 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…StarE (Galkin et al, 2020) leverages a message passing network, CompGCN (Vashishth et al, 2020), as an encoder to obtain the relation and entity embeddings, which are then fed into a transformer decoder to obtain the validity of facts. Hy-Transformer (Yu and Yang, 2021), GRAN and QUAD (Shomer et al, 2022) further improve it with alternative designs of encoders and via auxiliary training tasks. Relatively, these models, though useful, require a large number of parameters and are prone to overfitting.…”

Section: Key-value Pairsmentioning

confidence: 99%

“…Baselines We compare ShrinkE against various models, including m-TransH (Wen et al, 2016), RAE (Zhang et al, 2018), NaLP-Fix (Rosso et al, 2020), HINGE (Rosso et al, 2020), NeuInfer (Guan et al, 2020, BoxE (Abboud et al, 2020), Transformer andStarE (Galkin et al, 2020). Note that we exclude Hy-Transformer (Yu and Yang, 2021), GRAN and QUAD (Shomer et al, 2022) and 2) they leverage auxiliary training tasks, which can also be incorporated into our framework and we leave as one future work.…”

Section: Environments and Hyperparametersmentioning

confidence: 99%

“…Currently, the main goal of ShrinkE is to model inference patterns directly in the embedding space for hyper-relational KGs and we do not explore more advanced training strategies that have recently been proposed. For example, recent works (Yu and Yang, 2021;Shomer et al, 2022) have demonstrated that adding auxiliary training tasks, e.g., the task of predicting qualifier entities, can further improve the overall performance. We believe such auxiliary training tasks can also benefit ShrinkE and we leave it as future work.…”

Section: Limitationsmentioning

confidence: 99%

“…Link prediction is then achieved by modeling the validity of the primary triple and its compatibility with each annotated qualifier (Guan et al, 2020;Rosso et al, 2020). More complicated graph encoders and decoders (Galkin et al, 2020;Yu and Yang, 2021;Shomer et al, 2022) are proposed to further boost the performance. However, they require a relatively huge number of parameters that make them prone to overfitting.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Shrinking Embeddings for Hyper-Relational Knowledge Graphs

Xiong¹,

Nayyeri²,

Pan³

et al. 2023

Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

View full text Add to dashboard Cite

Link prediction on knowledge graphs (KGs) has been extensively studied on binary relational KGs, wherein each fact is represented by a triple. A significant amount of important knowledge, however, is represented by hyperrelational facts where each fact is composed of a primal triple and a set of qualifiers comprising a key-value pair that allows for expressing more complicated semantics. Although some recent works have proposed to embed hyper-relational KGs, these methods fail to capture essential inference patterns of hyperrelational facts such as qualifier monotonicity, qualifier implication, and qualifier mutual exclusion, limiting their generalization capability. To unlock this, we present ShrinkE, a geometric hyper-relational KG embedding method aiming to explicitly model these patterns. ShrinkE models the primal triple as a spatial-functional transformation from the head into a relation-specific box. Each qualifier "shrinks" the box to narrow down the possible answer set and, thus, realizes qualifier monotonicity. The spatial relationships between the qualifier boxes allow for modeling core inference patterns of qualifiers such as implication and mutual exclusion. Experimental results demonstrate ShrinkE's superiority on three benchmarks of hyper-relational KGs.

show abstract

Section: Key-value Pairsmentioning

confidence: 99%

Section: Environments and Hyperparametersmentioning

confidence: 99%

Section: Limitationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Shrinking Embeddings for Hyper-Relational Knowledge Graphs

Xiong¹,

Nayyeri²,

Pan³

et al. 2023

Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

View full text Add to dashboard Cite

show abstract

“…To address this issue, various temporal knowledge graph embedding (KGE) models are proposed to encode entities and relations in a low-dimensional space using translation-based functions [1,2], a deep neural networkbased method [3,4], and a tensor decomposition method [5,6]. As these methodologies have progressed, numerous strategies for TKG completion have been proposed, including geometric methods (ChronoR [7], BoxTE [8], TLT-KGE [9], RotateQVS [10], HTTR [11], PTKE [12]); tensor decomposition methods (QDN [13], TBDRI [14]); deep learning and embedding-based methods (TASTER [15], TeAST [16], RoAN [17], BiQCap [18]); and graph neural network-based reasoning methods (TARGCN [19], T-GAP [20], TAL-TKGC [21]). Interpolation is a statistical method that utilizes the relevant known values to estimate an unknown value or set [19].…”

Section: Introductionmentioning

confidence: 99%

A Multi-View Temporal Knowledge Graph Reasoning Framework with Interpretable Logic Rules and Feature Fusion

Xu,

Bao,

et al. 2024

Electronics

View full text Add to dashboard Cite

A temporal knowledge graph represents temporal information between entities in a multi-relational graph. Its reasoning aims to infer and predict potential links among entities. Predicting time-aware entities is a challenging task due to significant differences in entity appearances over time, such as different timestamps and frequencies. Current embedding-based similarity-matching methods have been introduced for predicting temporal facts. However, they lack deterministic logical explainability and cannot model the dynamic evolution of entities over time. To address these challenges, we propose a novel framework for temporal knowledge graph reasoning based on multi-view feature fusion (MVFF). First, MVFF extracts logical rules and uses the Gumbel-Softmax trick to sample high-quality rules. Second, it integrates logical rules, temporal quadruples, and factual triples to capture semantic features, temporal information, and structural information to solve link prediction tasks. Through experiments on four benchmark datasets, we show that MVFF outperforms state-of-the-art methods, providing not only better performance but also interpretable results.

show abstract