Time-aware Entity Alignment using Temporal Relational Attention

Xu, Chengjin; Su, Fenglong; Xiong, Bo; Lehmann, Jens

doi:10.1145/3485447.3511922

Cited by 18 publications

(29 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As the formats of storing temporal information in TKGs are almost identical, the alignment information can be obtained easily, which naturally enhances the EA performance. Nevertheless, recent studies [53,54] that applies TKGs to EA still suffer from the following two problems.…”

Section: G S G Tmentioning

confidence: 99%

“…They [53,54] follow studies that applies KGs to EA, which cannot perform EA until the prealigned seeds are obtained. However, unlike KGs, temporal information in relational triples of TKGs is naturally aligned since they represent real-world time points or time periods [53,54]. This character of TKGs provides the opportunity of developing time-aware EA methods in an unsupervised fashion by treating temporal information as seed alignment.…”

Section: G S G Tmentioning

confidence: 99%

“…• Insufficient temporal information discovery. They [53,54] simply create embeddings for each timestamp, and treat the time intervals as relation types to enhance the graph learning process. These processes highly rely on the seed alignment, where the temporal information has not been fully exploited, and thus the accuracy of EA is limited.…”

Section: G S G Tmentioning

confidence: 99%

“…We present an EA method for TKGs, DualMatch, that is able to address the above two problems. To tackle the first problem, instead of treating time intervals simply as relations [53,54], we model the temporal information using a learning-free encoder. Specifically, we derive pseudo seeds from the encoder to train GNN-based EA models.…”

Section: G S G Tmentioning

confidence: 99%

See 3 more Smart Citations

Unsupervised Entity Alignment for Temporal Knowledge Graphs

Liu¹,

Wu²,

Li³

et al. 2023

Preprint

View full text Add to dashboard Cite

Entity alignment (EA) is a fundamental data integration task that identifies equivalent entities between different knowledge graphs (KGs). Temporal Knowledge graphs (TKGs) extend traditional knowledge graphs by introducing timestamps, which have received increasing attention. State-of-the-art time-aware EA studies have suggested that the temporal information of TKGs facilitates the performance of EA. However, existing studies have not thoroughly exploited the advantages of temporal information in TKGs. Also, they perform EA by pre-aligning entity pairs, which can be labor-intensive and thus inefficient. In this paper, we present DualMatch that effectively fuses the relational and temporal information for EA. DualMatch transfers EA on TKGs into a weighted graph matching problem. More specifically, DualMatch is equipped with an unsupervised method, which achieves EA without necessitating the seed alignment. DualMatch has two steps: (i) encoding temporal and relational information into embeddings separately using a novel label-free encoder, Dual-Encoder; and (ii) fusing both information and transforming it into alignment using a novel graph-matching-based decoder, GM-Decoder. DualMatch is able to perform EA on TKGs with or without supervision, due to its capability of effectively capturing temporal information. Extensive experiments on three real-world TKG datasets offer the insight that DualMatch significantly outperforms the state-of-the-art methods.

show abstract

Section: G S G Tmentioning

confidence: 99%

Section: G S G Tmentioning

confidence: 99%

Section: G S G Tmentioning

confidence: 99%

Section: G S G Tmentioning

confidence: 99%

See 2 more Smart Citations

Unsupervised Entity Alignment for Temporal Knowledge Graphs

Liu¹,

Wu²,

Li³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Particularly, the time-aware entity alignment approach based on graph neural networks (TEA-GNN) (Xu et al, 2021) first designs a time-aware GNN to cope with TEA, which exploits a time-aware mechanism to introduce the time information into entity embeddings. The time-aware entity alignment using temporal relational attention (TREA) (Xu et al, 2022), on the other hand, incorporates temporal embeddings to enrich the entity embeddings and achieves state-of-the-art performance. Nevertheless, they cannot fully tackle the aforementioned challenges brought by TEA.…”

Section: Introductionmentioning

confidence: 99%

Active Temporal Knowledge Graph Alignment

Zhou

Zeng

et al. 2023

International Journal on Semantic Web and Information Systems

View full text Add to dashboard Cite

Entity alignment aims to identify equivalent entity pairs from different knowledge graphs (KGs). Recently, aligning temporal knowledge graphs (TKGs) that contain time information has aroused increasingly more interest, as the time dimension is widely used in real-life applications. The matching between TKGs requires seed entity pairs, which are lacking in practice. Hence, it is of great significance to study TKG alignment under scarce supervision. In this work, the authors formally formulate the problem of TKG alignment with limited labeled data and propose to solve it under the active learning framework. As the core of active learning is to devise query strategies to select the most informative instances to label, the authors propose to make full use of time information and put forward novel time-aware strategies to meet the requirement of weakly supervised temporal entity alignment. Extensive experimental results on multiple real-world datasets show that it is important to study TKG alignment with scarce supervision, and the proposed time-aware strategy is effective.

show abstract