Adversarial Perturbations for Joint Entity and Relation Extraction
The goal of the entity recognition and relation extraction task is to discover relational structures of entity mentions from unstructured texts. It is a central problem in information extraction since it is critical for tasks such as knowledge base population and question answering. In this work, we focus on extending the training procedure of our newly proposed general purpose joint model for entity recognition and relation extraction with adversarial training.
On January 2016, Giannis joined the IDLab research group at the Information Technology department of Ghent University, as a Ph.D. student. As a part of the UGent Text-to-Knowledge research cluster within IDLab, Giannis focused on Natural Language Processing techniques for extracting useful information from unstructured text and applied machine learning and neural network methods for core Natural Language Processing problems.