Delft University of Technology
Van Mourik Broekmanweg 6
I am a Postdoctoral Researcher in the Web Information Systems
group at Delft University of Technology, working on the Responsible Data Science track in the Digital Society
research programme. I am also a visiting researcher at the User Centric Data Science
group, Computer Science department at Vrije Universiteit Amsterdam.
I currently work on the FAIRView
project, a Google Digital News Innovation Fund
project, which focuses on generating adaptive news video summaries and the development of metrics to assess the quality of the video summaries and their potential for misinformation. To this end, we first investigate the usefulness of different types of video summaries in terms of how well they represent different aspects of the original video and the video as a whole. Second, we evaluate the video summaries in terms of fairness, accuracy, bias or misinformation. Recently, I received $5,000 in Google Cloud Platform
credits for this research project.
Before joining TU Delft on 1st of January 2019, I was a PhD student in the User Centric Data Science
group at Vrije Universiteit Amsterdam, under the supervision of prof.dr. Lora Aroyo
My research focused on understanding events and their semantic representation across various data types (text, image, video) and different domains (history, politics). This also includes understanding the role of events with respect to other types of entities, such as places, people, time expressions, as well as detecting their contextual relevance, salience or novelty. To realize this, I applied a number of computer science technologies, such as, natural language processing (NLP), semantic modeling, human computation, machine learning.
During my PhD I was awarded two times the IBM PhD Fellowship (2016-2017, 2017-2018).
I performed my research in the context of two projects:
CrowdTruth: novel approach for crowdsourcing ground truth data by harnessing the inter-annotator disagreement, i.e, representing the diversity of human interpretations, instead of enforcing agreement between annotators. The project was developed in collaboration with IBM.
DIVE+: a linked-data digital collection browser aimed at providing an integrated and interactive access to multimedia objects from various heterogeneous online collections. As part of the DIVE+ project I'm also involved in Clariah (Common Lab Research Infrastructure for the Arts and Humanities). In this research infrastructure, next to other media studies research tools, DIVE+ aims at supporting the media studies researchers and scholars by providing access to digital data and tools.