Abstract
Temporal knowledge graphs (TKGs) have been applied in many fields, reasoning over TKG which predicts future facts is an important task. Recent methods based on Graph Convolution Network (GCN) represent entities and relations in Euclidean space. However, Euclidean vectors cannot accurately distinguish entities in similar facts, it is necessary to further represent entities and relations in complex space. We propose Time-aware Quaternion Graph Convolution Network (T-QGCN) based on Quaternion vectors, which can more efficiently represent entities and relations in quaternion space to distinguish entities in similar facts. T-QGCN also adds a time-aware part to show the influence of the occurrence frequency of historical facts when reasoning. Specifically, T-QGCN uses QGCN with each historical fact frequency to aggregate graph structural information for each timestamp in TKGs and uses RNN to dynamically update entity representation and relation representation. To decode in quaternion space and better use historical representations, we design a new decoding module based on Convolution Neural Network (CNN) to help T-QGCN perform better. Extensive experiments show that T-QGCN has better performance than baselines for the entity prediction task on four datasets.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Trivedi, R., Farajtabar, M., Biswai, P., Zha, H.: Dyrep: Learning representations over dynamic graphs. In: ICLR (2019)
Wang, X., Li, X., Zhu, J., et al.: A local similarity-preserving framework for nonlinear dimensionality reduction with neural networks. In: DASFAA (2021)
Jin, W., Qu, M., Jin, X., Ren, X.: Recurrent event network: autoregressive structure inference over temporal knowledge graphs. In: EMNLP (2020)
Zhu, C., Chen, M., Fan, C., Cheng, G., Zhan, Y.: Learning from history: modeling temporal knowledge graphs with sequential copy-generation networks. In: ICLR (2021)
Li. Z., et al.: Temporal knowledge graph reasoning based on evolutional representation learning. In: SIGIR (2021)
Dai, Q.N., Tu, Q.N., Phuang D.: Quaternion graph neural network. In: ACML (2021)
Bordes, A., Usunier, N., Garcia, A., Weston, J., Yakhnenko, Q.: Translating embeddings for modeling multi-relational data. In: Advances in Neural Information Processing Systems (2013)
Wang, Z., Zhang, J., Feng, J., Chen, Z.: Knowledge graph embedding by translating on hyperplanes. In: AAAI (2014)
Lin, Y., Liu, Z., Sun, M., Lin, Y., Zhu, X.: Learning entity and relation embeddings for knowledge graph completion. In: AAAI (2015)
Nickel, M., Tresp, V., Kriegel, H.P.: A three-way model for collective learning on multi-relational data. In: ICML (2011)
Yang, B., Yih, W., He, X., Gao, J., Deng, L.: Embedding entities and relations for learning and inference in knowledge bases. In: ICLR (2015)
Sun, Z., Deng, Z.H., Nie, J.Y., Tang, J.: Rotate: Knowledge graph embedding by relational rotation in complex space. In: ICLR (2019)
Zhang, S., Tay, Y., Yao, L., Liu, Q.: Quaternion knowledge graph embedding. In: Neural Information Processing Systems (2019)
Dettmers, T., Minervini, P., Stenetorp, P., Riedel, S.: Convolutional 2d knowledge graph embeddings. In: AAAI (2018)
Kipf, T., Welling, M.: Semi-supervised classification with graph convolutional networks. In: Advances in Neural Information Processing Systems (2016)
Schlichtkrull, M., Kipf, T., Bloem, P., Berg, R., Titov, I., Welling, M.: Modeling relational data with graph convolutional networks. In: European Semantic Web Conference (2018)
Shang, C., Tang, Y., Huang, J., Bi, J., He, X., Zhou, B.: End-to-end structure-aware convolutional networks for knowledge base completion. In: AAAI (2019)
Garcia-Duran, A., Dumancic, S., Niepert, M.: Learning sequence encoders for temporal knowledge graph completion. In: EMNLP (2018)
Leblay, J., Chekol, M.W.: Deriving validity time in knowledge graph. In: International World Wide Web Conferences Steering Committee (2018)
Dasgupta, S., Ray, S.N., Talukdar, P.: Hyte: hyperplane based temporally aware knowledge graph embedding. In: EMNLP (2018)
Xu, C., Nayyeri, M., Alkhoury, F., Yazdi, H., Lehmann, J.: TeRo: a time-aware knowledge graph embedding via temporal rotation. In: International Conference on Computational Linguistics (2020)
Xu. C., Nayyeri, M., Alkhoury, F., Yazdi, H., Lehmann, J.: Temporal knowledge graph embedding on additive time series decomposition. In: International Semantic Web Conference (2020)
Trivedi, R., Dai, H., Wang, Y., Song, L.: Know-evolve: deep temporal reasoning for dynamic knowledge graphs. In: ICML (2017)
Boschee, E., et al.: Icews coded event data. In: Harvard dataverse (2015)
Mahdisoltani, F., Biega, J. A., Suchanek, F.: Yago3: a knowledge base from multilingual Wikipedia. In: CIDR (2014)
Wang, S., Cao, J., Yu, P.: Deep learning for spatio-temporal data mining: a survey. In: TKDE (2022)
Acknowledgement
This work is supported in part by the Major Key Project of PCL (Grant No. PCL2022A03).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Mo, C., Wang, Y., Jia, Y., Luo, C. (2023). Time-aware Quaternion Convolutional Network for Temporal Knowledge Graph Reasoning. In: Tanveer, M., Agarwal, S., Ozawa, S., Ekbal, A., Jatowt, A. (eds) Neural Information Processing. ICONIP 2022. Communications in Computer and Information Science, vol 1791. Springer, Singapore. https://doi.org/10.1007/978-981-99-1639-9_25
Download citation
DOI: https://doi.org/10.1007/978-981-99-1639-9_25
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-1638-2
Online ISBN: 978-981-99-1639-9
eBook Packages: Computer ScienceComputer Science (R0)