The paper explores the application of Knowledge Graph Embeddings (KGE) and Graph Neural Networks (GNN) in the context of the Web of Things (WoT). The focus is on how these techniques can help handle issues of noisy, messy or incomplete data in large-scale IoT/WoT knowledge graphs. The study investigates the performance of KGE and GNN methods on tasks such as link prediction and node classification. The results indicate promising performance of both methods on node classification, and superior performance of GNN approaches in link prediction.
Publication date: 24 Oct 2023
Project Page: Not provided
Paper: https://arxiv.org/pdf/2310.14866
