Structural Characteristics of Knowledge Graphs Determine the Quality of Knowledge Graph Embeddings Across Model and Hyperparameter Choices

Abstract

The realm of biomedicine is producing information at a rate far beyond the capacity of clinicians, researchers, and machine learning experts to analyse in full. Recently, developments in Knowledge Graphs (KGs) have facilitated the representation of all this information in an easily-integrable and easily-queryable format. With increasing academic and clinical interest in Knowledge Graph Em- beddings (KGEs), various KGE models have been developed to allow machine learning to efficiently run on these large Knowledge Graphs and predict new, previously unseen information about the domain. However, the need to validate hyperparameters for every new dataset, especially considering the time and expertise needed for validation and model training, have limited the use of KGEs in biology to those who have expertise in machine learning and knowledge engineering. This research presents a framework by which the effect of hyperparameters on model performance for a given KG can be modelled as a function of KG structure. The presented evaluation of the framework finds a clear effect of graph structure on hyperparameter fitness. This leads to the conclusion that more research into cross-dataset hyperparameter prediction and re-use holds promise for increasing the accessibility and usability of KGEs for biomedical applications.