Document Structure aware Relational Graph Convolutional Networks for Ontology Population
This addresses the challenge of improving ontology population for knowledge-based AI systems, though it appears incremental by incorporating document structure into an existing method.
The paper tackled the problem of populating ontologies by learning relationships between concepts in document corpora, achieving about 15 points higher accuracy than a stand-alone R-GCN model.
Ontologies comprising of concepts, their attributes, and relationships are used in many knowledge based AI systems. While there have been efforts towards populating domain specific ontologies, we examine the role of document structure in learning ontological relationships between concepts in any document corpus. Inspired by ideas from hypernym discovery and explainability, our method performs about 15 points more accurate than a stand-alone R-GCN model for this task.