Quaternion-Based Representation and Explicit Historical Enhancement for Temporal Knowledge Graph Completion
LÜ Jiaqi1, XIE Jun1, WANG Li2, WANG Dantong1, LUO Xiongyan1
1. College of Electronic Information Engineering, Taiyuan University of Technology, Jinzhong 030600; 2. College of Artificial Intelligence, Taiyuan University of Technology, Jinzhong 030600
Abstract：Existing temporal knowledge graph completion models struggle to effectively model the deep interactions between semantic information and multi-granularity temporal information and lack explicit the modeling of temporal sensitivity of historical information. To solve these problems, an approach for quaternion-based representation and explicit historical enhancement for temporal knowledge graph completion(QR-EH) is proposed. First, the semantic information of entities and relations, and the multi-granularity temporal information are mapped into the quaternion space. Deep interactions between semantics and time are achieved through Hamiltonian products. Then, an explicit historical retrieval module is designed. A temporal modulation mechanism is defined by this module. Time-sensitive historical patterns are accurately captured by weighting historical recurring events. Finally, an adaptive score fusion module is constructed. Global spatio-temporal information and historical recurring information are weighted and fused by this module. The final prediction results are generated. Experiments on three publicly available datasets demonstrate that QR-EH deeply mines temporal evolution and historical recurrence patterns and exhibits good generalization and interpretability.
[1] 何鹏,姚瑶,刘秋菊.时态知识图谱表示学习研究综述[J].计算机工程与应用, 2025, 61(14): 37-53. (He P, Yao Y, Liu Q J. Survey on temporal knowledge graph re-presentation learning. Computer Engineering and Applications, 2025, 61(14): 37-53.) [2] 许凯嘉,柳林,王海龙,等.时序知识图谱补全研究综述[J].计算机工程与应用, 2024, 60(22): 38-57. (Xu K J, Liu L, Wang H L, et al. Survey on temporal knowledge graph completion research. Computer Engineering and Applications, 2024, 60(22): 38-57.) [3] BaiI L Y, Xu L S, Zhu L. Question answering over temporal know-ledge graphs based on time sensitive graph neural network[J/OL]. Applied Soft Computing, 2025, 177. https://doi.org/10.1016/j.asoc.2025.113257. [4] Wang M J, Huo Y F, Zheng J H, et al. SC-TKGR: temporal know-ledge graph-based GNN for recommendations in supply chains[J/OL]. Electronics, 2025, 14(2). https://doi.org/10.3390/electronics14020222. [5] Jia N N, Yao C Y. A brief survey on deep learning-based temporal knowledge graph completion[J/OL]. Applied Sciences, 2024, 14(19). https://doi.org/10.3390/app14198871. [6] Meng X Q, Chen M, Pan Y. Temporal knowledge graph completion based on temporal-aware encoding and entity attention contrast[J/OL]. International Journal of Computational Intelligence Systems, 2025, 18(1). https://doi.org/10.1007/s44196-025-01047-4. [7] Jiang T S, Liu T Y, Ge T, et al. Encoding temporal information for time-aware link prediction[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing. Stroudsburg, USA: ACL, 2016: 2350-2354. [8] Lacroix T, Obozinski G, Usunier N. Tensor decompositions for temporal knowledge base completion[EB/OL]. [2026-02-21]. https://arxiv.org/pdf/2004.04926. [9] Wu J P, Cao M, Cheung J C K, et al. TeMP: temporal message passing for temporal knowledge graph completion[C]//Proceedings of the Conference on Empirical Methods in Natural Language Proce-ssing. Stroudsburg, USA: ACL, 2020: 5730-5746. [10] Dong H, Wang P Y, Xiao M,et al. Temporal inductive path neural network for temporal knowledge graph reasoning[J/OL]. Artificial Intelligence, 2024, 329. https://doi.org/10.1016/j.artint.2024.104085. [11] García-durán A, Dumančić S, Niepert M. Learning sequence encoders for temporal knowledge graph completion[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing. Stroudsburg, USA: ACL, 2018: 4816-4821. [12] Xu C J, Nayyeri M, Alkhoury F, et al. Temporal knowledge graph completion based on time series gaussian embedding[C]//Proceedings of the 19th International Semantic Web Conference. Berlin, Germany: Springer, 2020: 654-671. [13] Jin W, Qu M, Jin X S, et al. Recurrent event network: autoregressive structure inference over temporal knowledge graphs[C]//Proceedings of the Conference on Empirical Methods in Natural Language Processing. Stroudsburg, USA: ACL, 2020: 6669-6683. [14] Dasgupta S S, Ray S N, Talukdar P. HyTE: hyperplane-based temporally aware knowledge graph embedding[C]//Proceedings of the Conference on Empirical Methods in Natural Language Proce-ssing. Stroudsburg, USA: ACL, 2018: 2001-2011. [15] Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780. [16] Geng R S, Luo C C. Learning temporal granularity with quadruplet networks for temporal knowledge graph completion[J/OL]. Scientific Reports, 2025, 15(1). https://doi.org/ 10.1038/s41598-025-00446-z. [17] Zhang J C, Wan T Q, Mu C, et al. Learning granularity representation for temporal knowledge graph completion[C]//Proceedings of the 31st International Conference on Neural Information Proce-ssing. Cambridge, USA: MIT Press, 2025: 91-105. [18] Schlichtkrull M, Kipf T N, Bloem P, et al. Modeling relational data with graph convolutional networks[C]//Proceedings of the 15th International Conference on Semantic Web. Berlin, Germany: Springer, 2018: 593-607. [19] Zaremba W, Sutskever I, Vinyals O. Recurrent neural network regularization[EB/OL]. [2026-02-21]. https://arxiv.org/pdf/1409.2329. [20] Zhu C C, Chen M H, Fan C J, et al. Learning from history: mode-ling temporal knowledge graphs with sequential copy-generation networks[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(5): 4732-4740. [21] Ouyang G G, Lü A, Chen Y, et al. RLGNet: repeating-local-glo-bal history network for temporal knowledge graph reasoning[C]//Proceedings of the 28th International Conference on Computer Su-pported Cooperative Work in Design. Washington, USA: IEEE, 2025: 2310-2317. [22] Hamilton W R. On quaternions, or on a new system of imaginaries in algebra[M/OL]. [2026-02-21]. https://www.maths.tcd.ie/pub/HistMath/People/Hamilton/OnQuat/OnQuat.pdf. [23] Zhang S, Tay Y, Yao L N, et al. Quaternion knowledge graph embeddings[C]//Proceedings of the 33rd International Conference on Neural Information Processing Systems. Cambridge, USA: MIT Press, 2019: 2735-2745. [24] Chen K, Wang Y, Li Y T, ,et al. RotateQVS: representing temporal information as rotations in quaternion vector space for temporal knowledge graph completion[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics: Long Papers. Stroudsburg, USA: ACL, 2022, I: 5843-5857. [25] Zhang F W, Zhang Z, Ao X, et al. Along the time: timeline-traced embedding for temporal knowledge graph completion[C]//Proceedings of the 31st ACM International Conference on Information and Knowledge Management. New York, USA: ACM, 2022: 2529-2538. [26] Yu R G, Liu T, Yu J, ,et al. Combination of translation and rotation in dual quaternion space for temporal knowledge graph completion[C/OL]//Proceedings of the International Joint Conference on Neural Networks. Washington, USA: IEEE, 2023. https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10191552. [27] Guo J J, Zhao M K, Yu J, et al. EHPR: learning evolutionary hie-rarchy perception representation based on quaternion for temporal knowledge graph completion[J/OL]. Information Sciences, 2025, 688. https://doi.org/10.1016/j.ins.2024.121409. [28] Yu R G, Guo J Z, Zhao M K, ,et al. MTE: multi transformation of entities in quaternion vector space for temporal knowledge graph completion[C/OL]//Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing. Washington, USA: IEEE, 2025. https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10890169. [29] 张亚睿,谢珺,吕佳琪,等.基于Transformer和强化学习的时序知识图谱补全方法[J]. 小型微型计算机系统, 2026, 47(5): 1108-1126. (Zhang Y R, Xie J, Lü J Q, et al. Temporal knowledge graph completion method based on Transformer and reinforcement learning[J]. Journal of Chinese Computer Systems, 2026, 47(5): 1108-1126. [30] Li Y J, Sun S L, Zhao J. TiRGN: time-guided recurrent graph network with local-global historical patterns for temporal knowledge graph reasoning[C]//Proceedings of the 31st International Joint Conference on Artificial Intelligence Main Track. San Francisco, USA: IJCAI, 2022: 2152-2158. [31] Wang K Z, Han C R, Poon J. Re-Temp: Relation-aware temporal representation learning for temporal knowledge graph completion[C]//Proceedings of the Findings of the Association for Computational Linguistics. Stroudsburg, USA: ACL, 2023: 258-269. [32] Zhang F W, Zhang Z, Zhuang F Z, et al. Temporal knowledge graph reasoning with dynamic memory enhancement[J]. IEEE Transactions on Knowledge and Data Engineering, 2024, 36(11): 7115-7128. [33] 贾凯,王阳萍,杨景玉,等.融合局部-全局历史模式与历史知识频率的时序知识图谱补全方法[J].计算机应用研究, 2025, 42(6): 1727-1733. (Jia K, Wang Y P, Yang J Y, et al. Time-series knowledge graph completion method combining local-global historical pattern and historical knowledge frequency[J]. Application Research of Computers. 2025, 42(6): 1727-1733.) [34] Han Z, Chen P, Ma Y P, et al. xERTE: explainable reasoning on temporal knowledge graphs for forecasting future links[EB/OL].[2026-02-21]. https://arxiv.org/pdf/2012.15537. [35] Hu J, Zhu Y L, Teng F, et al. Temporal knowledge graph reaso-ning based on relation graphs and time-guided attention mechanism[J/OL]. Knowledge-Based Systems, 2024, 301. https://doi.org/10.1016/j.knosys.2024.112280. [36] Zhang J C, Hui B, Mu C, et al. Learning multi-graph structure for temporal knowledge graph reasoning[J/OL]. Expert Systems with Applications, 2024, 255. https://doi.org/10.1016/j.eswa.2024.124561. [37] Liu W T, Hasikin K, Khairuddin A S M, ,et al. A temporal know-ledge graph reasoning model based on recurrent encoding and con-trastive learning[J/OL]. PeerJ Computer Science, 2025, 11. https://pdf.semanticscholar.org/3bbf/695f23f8c4ee12dbadd50088b89bfd1a8a31.pdf.
2026, Vol. 39 
