Dihedron Algebraic Embeddings for Spatio-Temporal Knowledge Graph Completion (Research Colloquium presentation by Umair Qudus)
3 months ago
Dihedron Algebraic Embeddings for Spatio-Temporal Knowledge Graph Completion
Umair Qudus - ESR 7, University of Paderborn
Many knowledge graphs (KG) contain spatial and temporal information. Most KG embedding models follow triple-based representation and often neglect the simultaneous consideration of the spatial and temporal aspects. Encoding such higher dimensional knowledge necessitates the consideration of true algebraic and geometric aspects. Hypercomplex algebra provides the foundation of a well defined mathematical system among which the Dihedron algebra with its rich framework is suitable to handle multidimensional knowledge. In this paper, authors propose an embedding model that uses Dihedron algebra for learning such spatial and temporal aspects. The evaluation results show that their model performs significantly better than other adapted models.
The slides of the presentation can be found here.