On the representation and embedding of knowledge bases beyond binary relations
arXiv:1604.08642
Abstract
The models developed to date for knowledge base embedding are all based on the assumption that the relations contained in knowledge bases are binary. For the training and testing of these embedding models, multi-fold (or n-ary) relational data are converted to triples (e.g., in FB15K dataset) and interpreted as instances of binary relations. This paper presents a canonical representation of knowledge bases containing multi-fold relations. We show that the existing embedding models on the popular FB15K datasets correspond to a sub-optimal modelling framework, resulting in a loss of structural information. We advocate a novel modelling framework, which models multi-fold relations directly using this canonical representation. Using this framework, the existing TransH model is generalized to a new model, m-TransH. We demonstrate experimentally that m-TransH outperforms TransH by a large margin, thereby establishing a new state of the art.
arXiv comment8 pages, to appear in IJCAI 2016
References in corpus (3)
Cited by in corpus (10)
- A Comprehensive Survey of Graph Embedding: Problems, Techniques and Applications
- Representation Learning for Dynamic Graphs: A Survey
- BoxE: A Box Embedding Model for Knowledge Base Completion
- HAHE: Hierarchical Attention for Hyper-Relational Knowledge Graphs in Global and Local Level
- Representation Learning on Hyper-Relational and Numeric Knowledge Graphs with Transformers
- DHGE: Dual-View Hyper-Relational Knowledge Graph Embedding for Link Prediction and Entity Typing
- Knowledge Hypergraph Embedding Meets Relational Algebra
- NQE: N-ary Query Embedding for Complex Query Answering over Hyper-Relational Knowledge Graphs
- RDF-star2Vec: RDF-star Graph Embeddings for Data Mining
- Searching to Sparsify Tensor Decomposition for N-ary Relational Data