Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Temporal Knowledge Graph Completion Based on Time Series Gaussian Embedding

: Xu, Chenjin; Nayyeri, Mojtaba; Alkhoury, Fouad; Yazdi, Hamed; Lehmann, Jens


Pan, J.Z.:
The Semantic Web - ISWC 2020. 19th International Semantic Web Conference. Proceedings. Pt.I : Athens, Greece, November 2-6, 2020
Cham: Springer Nature, 2020 (Lecture Notes in Computer Science 12506)
ISBN: 978-3-030-62418-7 (Print)
ISBN: 978-3-030-62419-4 (Online)
International Semantic Web Conference (ISWC) <19, 2020, Online>
European Commission EC
H2020; 812997; Cleopatra
Bundesministerium für Bildung und Forschung BMBF (Deutschland)
01IS18050F; MLWin
Fraunhofer IAIS ()
temporal knowledge graph; knowledge representation and reasoning; time series decomposition

Knowledge Graph (KG) embedding has attracted more attention in recent years. Most KG embedding models learn from time-unaware triples. However, the inclusion of temporal information besides triples would further improve the performance of a KGE model. In this regard, we propose ATiSE, a temporal KG embedding model which incorporates time information into entity/relation representations by using Additive Time Series decomposition. Moreover, considering the temporal uncertainty during the evolution of entity/relation representations over time, we map the representations of temporal KGs into the space of multi-dimensional Gaussian distributions. The mean of each entity/relation embedding at a time step shows the current expected position, whereas its covariance (which is temporally stationary ) represents its temporal uncertainty. Experimental results show that ATiSE significantly outperforms the state-of-the-art KGE models and the existing temporal KGE models on link prediction over four temporal KGs.