Isabelle van der Vegt
Bio
I am a PhD student at the UCL Department of Security and Crime Science. I also work as a research assistant on the ERC-funded GRIEVANCE project. The focus of my research is understanding threats of violence by using computational linguistics. I am interested in developing solutions to better understand and counter radicalisation, extremism, terrorism, and other forms of violence.
News (spring/summer 2019)
- We have conducted interviews with social media companies for our project on human-machine integration for extremist content removal. The project is funded by RUSI (with Paul Gill, Stuart Macdonald, and Bennett Kleinberg and a report will be published on their website shortly.
- I will speak at the Society for Terrorism Research conference in Oslo, 20-21 June 2019. My talk will focus on developing the GRIEVANCE dictionary: understanding terrorist language use (with Bennett Kleinberg, Paul Gill). I am also a co-author for the paper presentation: Towards Open and Reproducible Terrorism Studies (with Sandy Schumann, Bart Schuurman, & Paul Gill).
- I will speak at the Terrorism & Social Media conference in Swansea, 25-26 June 2019. The talk will focus on linguistic trajectories of YouTube’s alt-right (with Maximilian Mozes, Bennett Kleinberg, Paul Gill).
- I will present at the POLTEXT conference, on 14-15 September in Tokyo, Japan. The talk will discuss the forthcoming paper: ‘Online influence, offline violence: Linguistic responses to the ‘Unite the Right’ rally (with Maximilian Mozes, Bennett Kleinberg, Paul Gill).
Brief CV
- 2018-present: PhD student UCL Department of Security & Crime Science, supervised by Paul Gill & Bennett Kleinberg
- 2015-2017: Research Master’s degree in Psychology, University of Amsterdam
- 2012-2015: Bachelor of Liberal Arts & Sciences (Psychology + Linguistics), University College Utrecht
Recent publications & pre-prints
- Soldner, F., Ho, J., Makhortykh, M., van der Vegt, I., Mozes, M., Kleinberg, B. (2019). Uphill from here: Sentiment patterns in videos from left- and right-wing YouTube news channels. 3rd Workshop on Natural Language Processing and Computational Social Science. Paper.
- Kleinberg, B., van der Vegt, I., Arntz, A., Verschuere, B. (pre-print). Detecting deceptive communication through linguistic concreteness. Paper here.
- Kleinberg, B., Mozes, M., & van der Vegt, I. (2018). Identifying the sentiment styles of YouTube’s vloggers. Proceedings of EMNLP 2018. Click here for the paper, data, and code.
Previous activities
- At the 2019 UCL Open Science Day, we spoke about our open science seminar series at the Department of Security and Crime Science. You can find the website for JDI Open here and our slides on ‘Promoting Open (Crime) Science’ here.
- I co-organised a workshop on linguistic temporal trajectory analyis at the European Computational Social Science symposium 2018. You can find more information here.
- Violent mood swings: intra-textual sentiment of lone-actor manifestos. Talk at VOX-POL: Violent extremism, terrorism, and the internet: present and future trends, Amsterdam, Netherlands (August 21, 2018). Slides and data.
- Violent mood swings: intra-textual sentiment of violent lone actor manifestos. Talk at the annual Society for Terrorism Research conference, Liverpool, UK (September 6, 2018).