# Time-aware Graph Neural Networks for Entity Alignment between Temporal Knowledge Graphs Source: [Time-aware Graph Neural Networks for Entity Alignment between Temporal Knowledge Graphs]() Collection index: [llms.txt]() Data updated: 2025-10-28T13:56:52.309Z Authors: Chengjin Xu, Fenglong Su, Jens Lehmann Published: 2022-03-04T06:41:51.000Z ## Abstract Entity alignment aims to identify equivalent entity pairs between different knowledge graphs (KGs). Recently, the availability of temporal KGs (TKGs) that contain time information created the need for reasoning over time in such TKGs. Existing embedding-based entity alignment approaches disregard time information that commonly exists in many large-scale KGs, leaving much room for improvement. In this paper, we focus on the task of aligning entity pairs between TKGs and propose a novel Time-aware Entity Alignment approach based on Graph Neural Networks (TEA-GNN). We embed entities, relations and timestamps of different KGs into a vector space and use GNNs to learn entity representations. To incorporate both relation and time information into the GNN structure of our model, we use a time-aware attention mechanism which assigns different weights to different nodes with orthogonal transformation matrices computed from embeddings of the relevant relations and timestamps in a neighborhood. Experimental results on multiple real-world TKG datasets show that our method significantly outperforms the state-of-the-art methods due to the inclusion of time information. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Entity Alignment on DICEWS-200 Less Seed Setting (test)](): Entity Alignment; DICEWS-200 Less Seed Setting (test); H@1: 87.6. - [Entity Alignment on DICEWS unobserved entity pairs open-world (test)](): Entity Alignment; DICEWS unobserved entity pairs open-world (test); H@1: 15.5. - [Entity Alignment on DICEWS observed entity pairs open-world (test)](): Entity Alignment; DICEWS observed entity pairs open-world (test); H@1: 0.392. - [Entity Alignment on DICEWS-1K Normal Setting (10% seeds) (test)](): Entity Alignment; DICEWS-1K Normal Setting (10% seeds) (test); H@1: 88.7. - [Entity Alignment on WY50K-1K Less Seed Setting (test)](): Entity Alignment; WY50K-1K Less Seed Setting (test); H@1: 72.3. - [Entity Alignment on WY50K-5K Normal Setting (10% seeds) (test)](): Entity Alignment; WY50K-5K Normal Setting (10% seeds) (test); H@1: 87.9. - [Temporal Entity Alignment on YAGO-WIKI180K Dense-tem (6.6%) 2000 seeds](): Temporal Entity Alignment; YAGO-WIKI180K Dense-tem (6.6%) 2000 seeds; MRR: 84.3. - [Temporal Entity Alignment on BETA (30% seeds)](): Temporal Entity Alignment; BETA (30% seeds); H@1: 60.9. - [Temporal Entity Alignment on BETA (10% seeds)](): Temporal Entity Alignment; BETA (10% seeds); Hits@1: 55.7. - [Temporal Entity Alignment on DICEWS 1K](): Temporal Entity Alignment; DICEWS 1K; MRR: 91.1. - [Temporal Entity Alignment on DICEWS 200](): Temporal Entity Alignment; DICEWS 200; MRR: 0.902. - [Temporal Entity Alignment on YAGO-WIKI180K 2000 seeds (All)](): Temporal Entity Alignment; YAGO-WIKI180K 2000 seeds (All); MRR: 35.2. - [Temporal Entity Alignment on YAGO-WIKI180K Non-tem (50.4%) 2000 seeds](): Temporal Entity Alignment; YAGO-WIKI180K Non-tem (50.4%) 2000 seeds; MRR: 0.145. - [Temporal Entity Alignment on YAGO-WIKI180K Sparse-tem (43%) 2000 seeds](): Temporal Entity Alignment; YAGO-WIKI180K Sparse-tem (43%) 2000 seeds; MRR: 51.9. - [Temporal Entity Alignment on YAGO-WIKI50K 5K](): Temporal Entity Alignment; YAGO-WIKI50K 5K; MRR: 90.9. - [Temporal Entity Alignment on YAGO-WIKI50K 1K](): Temporal Entity Alignment; YAGO-WIKI50K 1K; MRR: 77.5. ## Original sources and implementation [Original paper]() [PDF]()