Document Structure Aware Relational Graph Convolutional Networks for Ontology Population
Abstract
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.
Cite
Text
Shalghar et al. "Document Structure Aware Relational Graph Convolutional Networks for Ontology Population." ICLR 2022 Workshops: DLG4NLP, 2022.Markdown
[Shalghar et al. "Document Structure Aware Relational Graph Convolutional Networks for Ontology Population." ICLR 2022 Workshops: DLG4NLP, 2022.](https://mlanthology.org/iclrw/2022/shalghar2022iclrw-document/)BibTeX
@inproceedings{shalghar2022iclrw-document,
title = {{Document Structure Aware Relational Graph Convolutional Networks for Ontology Population}},
author = {Shalghar, Abhay M and Kumar, Ayush and Ganesan, Balaji and Kannan, Aswin and Parekh, Akshay and G, Shobha},
booktitle = {ICLR 2022 Workshops: DLG4NLP},
year = {2022},
url = {https://mlanthology.org/iclrw/2022/shalghar2022iclrw-document/}
}