Research Fellow

Georgeta Bordea  is a postdoctoral researcher in Natural Language Processing that combines her industry and academic experience to push the limits of information extraction from text by envisioning domain-adaptive algorithms that are applied from newsrooms and bookstores to human resource departments and marketing agencies. She received her PhD in 2014 and has over 20 peer-reviewed publications since 2010, including 2 journal articles and 11 refereed conference publications, with an h-index of 8. Her research has attracted funding from national agencies including Science Foundation Ireland, Enterprise Ireland, and the Agence nationale de la recherche, but also from initiatives funded by the European Commission including the PROMISE Network of Excellence. Among her contributions she is most proud of setting up an evaluation framework for taxonomy extraction from text and creating one of the first multilingual datasets for this task. This enabled measurable progress in the field, allowing multiple research groups to compare their approaches as part of two consecutive SemEval shared tasks that she recently organised. She was one of the main contributors of Saffron, a system that automatically extracts topical hierarchies from text for visual browsing and exploration of large text collections. Georgeta Bordea has broad experience in analysing corpora from different sources and domains including enterprise documents, scientific publications, news articles, fiction books, and monographs. Her long term goal is to to enhance real-world applications with a semantic understanding​ of​ words.