Abstract The existing temporal knowledge graph representation methods cannot capture the complex relationships within quadruple well. Most of the neural network based models are unable to model time-varying knowledge and capture rich feature information. Moreover, the interaction between entities and relations in these models is poor. Therefore, a multi-scale dilated convolutional neural network model based on attention mechanism(MSDCA) is proposed. Firstly, a time-aware relation representation is obtained using long short-term memory. Secondly, a multi-scale dilated convolutional neural network is employed to improve the interactivity of the quadruple. Finally, a multi-scale attention mechanism is utilized to capture critical features to improve completion ability of MSDCA. Link prediction experiments on multiple public temporal datasets show the superiority of MSDCA.
Fund:National Natural Science Foundation of China(No.61672159), Project in Industry-University Cooperation of Fujian
Corresponding Authors:
WANG Jingbin, master, associate professor. Her research interests include knowledge graph, relation reasoning, distributed data management and knowledge representation.
About author:: LAI Xiaolian, master student. Her research interests include knowledge graph, relation reasoning and knowledge representation. LEI Jing, master student. Her research interests include knowledge graph, relation reasoning and knowledge representation. ZHANG Jingxuan, master student. Her research interests include knowledge graph, relation reasoning and knowledge representation.
[1] BOSCHEE E, LAUTENSCHLAGER J, O'BRIEN S, et al. ICEWS Coded Event Data[M/OL]. [2021-01-22]. https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/28075. [2] LEETARU K, SCHRODT P A. GDELT: Global Data on Events, Location, and Tone, 1979-2012[C/OL]. [2021-01-22]. http://citeseerx.ist.psu.edu/viewdoc/download;jsessionid=08494061C6CC8ADB678F4223F56689BB?doi=10.1.1.686.6605&rep=rep1&type=pdf. [3] ERXLEBEN F, GÜNTHER M, KRÖTZSCH M, et al. Introducing Wikidata to the Linked Data Web // Proc of the International Semantic Web Conference. Berlin, Germany: Springer, 2014: 50-65. [4] MAHDISOLTANI F, BIEGA J, SUCHANEK F M. YAGO3: A Knowledge Base from Multilingual Wikipedias[C/OL]. [2021-01-13]. https://suchanek.name/work/publications/cidr2015.pdf. [5] BORDES A, USUNIER N, GARCIA-DURAN A, et al. Translating Embeddings for Modeling Multi-relational Data // Proc of the 26th International Conference on Neural Information Processing Systems. Cambridge, USA: The MIT Press, 2013, II: 2787-2795. [6] YANG B S, YIH W, HE X D, et al. Learning Multi-relational Semantics Using Neural-Embedding Models[C/OL]. [2021-01-13]. https://arxiv.org/pdf/1411.4072.pdf. [7] KAZEMI S M, POOLE D. SimpLE Embedding for Link Prediction in Knowledge Graphs // Proc of the 32nd International Conference on Neural Information Processing. Berlin, Germany: Springer, 2018: 4289-4300. [8] SUN Z Q, DENG Z H, NIE J Y, et al. Rotate: Knowledge Graph Embedding by Relational Rotation in Complex Space[C/OL]. [2021-01-13]. https://arxiv.org/pdf/1902.10197.pdf. [9] DETTMERS T, MINERVINI P, STENETORP P, et al. Convolutional 2D Knowledge Graph Embeddings[C/OL]. [2021-01-13]. https://arxiv.org/pdf/1707.01476.pdf. [10] VASHISHTH S, SANYAL S, NITIN V, et al. InteractE: Improving Convolution-Based Knowledge Graph Embeddings by Increa-sing Feature Interactions[C/OL]. [2021-01-13]. https://arxiv.org/pdf/1911.00219.pdf. [11] JIANG T S, LIU T Y, GE T, et al. Towards Time-Aware Know-ledge Graph Completion // Proc of the 26th International Confe-rence on Computational Linguistics(Technical Papers). Stroudsburg, USA: ACL, 2016: 1715-1724. [12] DASGUPTA S S, RAY S N, TALUKDAR P. HyTE: Hyperplane-Based Temporally Aware Knowledge Graph Embedding // Proc of the Conference on Empirical Methods in Natural Language Pro-cessing. Stroudsburg, USA: ACL, 2018: 2001-2011. [13] GARCÍA-DURÁN A, DUMANČIĆ S, NIEPERT M. Learning Sequence Encoders for Temporal Knowledge Graph Completion[C/OL]. [2021-01-13]. https://arxiv.org/pdf/1809.03202.pdf. [14] GOEL R, KAZEMIi S M, BRUBAKER M, et al. Diachronic Embedding for Temporal Knowledge Graph Completion[C/OL]. [2021-01-13]. https://arxiv.org/pdf/1907.03143.pdf. [15] XU C J, NAYYERI M, ALKHOURY F, et al. TeRo: A Time-Aware Knowledge Graph Embedding via Temporal Rotation // Proc of the 28th International Conference on Computational Linguistics. Berlin, Germany: Springer, 2020: 1583-1593. [16] LI X, WANG W H, HU X L, et al. Selective Kernel Networks // Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2019: 510-519. [17] CHEN L C, PAPANDREOU G, KOKKINOS I, et al. Deeplab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 40(4): 834-848. [18] TRIVEDI R, DAI H J, WANG Y C, et al. Know-Evolve: Deep Temporal Reasoning for Dynamic Knowledge Graphs // Proc of the 34th International Conference on Machine Learning. Sydney, Australia: PMLR Press, 2017: 3462-3471. [19] BORDES A, WESTON J, COLLOBERT R, et al. Learning Structured Embeddings of Knowledge Bases // Proc of the 25th AAAI Conference on Artificial Intelligence. Palo Alto, USA: AAAI Press, 2011: 301-306.
2021, Vol. 34 
