Researchers from the ADAPT Centre and University of Galway will present their work at the SEMANTiCS conference in Vienna on the 5th September. John McCrae (also of the Insight Centre for Data Analytics) and Shubhanker Banerjee, who are affiliated with both the Research Ireland ADAPT Centre and the University of Galway, along with Bharathi Raja Chakravarthi from the University of Galway, co-authored the paper DA-ATE: Data Augmentation for Automatic Term Extraction.
The research addresses the problem of limited annotated corpora for training deep learning models in automatic term extraction across different domains. Using large language models for data augmentation, the team carried out experiments to evaluate the impact of augmentation at the context and term levels. They note that performance improvements across multiple domains further establishes the method’s effectiveness, as described in the paper.
For more information about the conference, visit the website here.
