Yesterday, I successfully defended his PhD on Extracting Fine-Grained Event and Sentiment from Economic News. I was supervised by prof. Véronique Hoste, prof. Els Lefever and prof. Diane Breesch. The actual defence was preceded by a workshop on “NLP for Economics” with around 40 people attending.
The past years I have worked on machine learning-based approaches for obtaining structured factual data alongside subjective information from business news for use in financial applications. To this purpose I created an extensive dataset for fine-grained event extraction and sentiment analysis in economic news, named SENTiVENT. This novel resource was validated with machine learning experiments in which state-of-the-art deep learning models were applied in coarse- and fine-grained settings.
Front cover of dissertation/thesis Extracting Fine-Grained Events and Sentiment from Economic News.
You can obtain my dissertation at the following preprint servers:
Below is a recording of the public defense.
And here are some photos.
Received the baret and giving the concluding speech.
Presenting my dissertation.