Embedding based Link Prediction for Knowledge Graph Completion
Biswas, Russa
Abstract:
This thesis proposes a novel Knowledge Graph (KG) embedding model for Link Prediction (LP) for Knowledge Graph Completion (KGC). The missing links in a KG are predicted based on the
existing contextual information as well as textual entity descriptions.
The model outperforms the state-of-the-art (SOTA) model DKRL for
FB15k and FB15k-237 datasets.
existing contextual information as well as textual entity descriptions.
The model outperforms the state-of-the-art (SOTA) model DKRL for
FB15k and FB15k-237 datasets.
| Zugehörige Institution(en) am KIT | Institut für Angewandte Informatik und Formale Beschreibungsverfahren (AIFB) |
| Publikationstyp | Proceedingsbeitrag |
| Publikationsmonat/-jahr | 08.2020 |
| Sprache | Englisch |
| Identifikator | KITopen-ID: 1000134456 |
| Erschienen in | Proceedings of the 1st Doctoral Consortium at the European Conference on Artificial Intelligence (DC-ECAI 2020). Hrsg.: J. M. Alonso Moral |
| Veranstaltung | 1st Doctoral Consortium at the European Conference on Artificial Intelligence (2020), Santiago de Compostela, Spanien, 29.08.2020 – 30.08.2020 |
| Verlag | EC-Libros e capı́tulos de libros |
| Seiten | 69-70 |
| Externe Relationen | Siehe auch |