20.500.12556/DKUM-87097 Knowledge Graph Completion with Triple Structure and Text Representation Knowledge Graphs (KGs) describe objective facts in the form of RDF triples, each triple contains sufficient semantic information and triple structure information. Knowledge Graph Completion (KGC) is to acquire new knowledge by predicting hidden relationships between entities and adding the new knowledge to the KG. At present, the mainstream KGC approaches only applied the triple structure information or only utilized the semantic information of the text. This paper proposes an approach (TSTR) using BERT and deep neural networks to fully extract the semantic information of knowledge, and designs an aggregated re-ranking scheme that incorporates existing graph embedding approach to learn the structural information of triples. In experiments, the approach achieves state-of-the-art performance on three benchmark datasets, and outperforms recent KGC approaches on sparsely connected datasets. knowledge graph completion BERT deep convolutional architecture re-ranking globoka konvolucijska arhitektura grafi znanja BERT procesiranje naravnih jezikov true false true Springer (Atlantis) Angleški jezik Slovenski jezik https://link.springer.com/journal/44196 © The Author(s) 2023 Članek v reviji 2024-02-19 08:53:49 2024-02-19 09:54:55 2024-06-13 07:45:40 0000-00-00 00:00:00 2023 0 0 Strr. 1-12 Št. članka 95 Letn. 16 Published: 30 May 2023 0000-00-00 Zaloznikova Objavljeno NiDoloceno 2023-01-15 2023-05-11 2023-05-30 004.9 162928643 10.1007/s44196-023-00271-0 1875-6883 Liu-2023-Knowledge_Graph_Completion_with_Tripl.pdf Liu-2023-Knowledge_Graph_Completion_with_Tripl.pdf 1 6D6B12816525821AD6FB69DA39E97446 2933678d7285a38da3d84b5a08ba429972d3c2df10ec4553e586609d90f46451 bcae1669-cf02-11ee-afa3-00155d000105 https://dk.um.si/Dokument.php?lang=slv&id=176247 https://link.springer.com/article/10.1007/s44196-023-00271-0 0 c8fcb212-cf02-11ee-afa3-00155d000105 https://dk.um.si/Dokument.php?lang=slv&id=176248 Fakulteta za elektrotehniko, računalništvo in informatiko 0 0 0