Loading Events

« All Events

  • This event has passed.

Symposium “NLP for the Vaccination Debate”

June 19th, 2018 @ 12:00 am

You are invited to join us for the symposium “NLP for the Vaccination Debate” on Wednesday 27 June, organised by the CLTL in collocation with the PhD defense of Isa Maks. We will present our richly-annotated corpus on the vaccination debate, announce its release and discuss some of the recent developments in Natural Language processing and their application to the online vaccination debate. The invited speakers are Sabine Bergler and Antal van den Bosch.


Symposium “NLP for the Vaccination Debate”
Held in collocation with the PhD defense of Isa Maks

When:Wednesday 27 June 2018 (13:00 – 15:00)
Where: Vrije Universiteit Amsterdam, room Agora-2 (main building)

More information: http://www.cltl.nl/event/symposium-nlp-for-the-vaccination-debate

Registration: https://goo.gl/forms/2Y8Ltgz5WZaFKV803


Sabine Bergler (Concordia University)
Robust versus volatile vs transient textual features

Drawing on different data sources, I will demonstrate how the robust inherent language features help to localize volatile features that are specific to domain, task, or corpus annotation standards. I will argue that especially in the quickly expanding semi-automatic text based research between social sciences and humanities, where the discovery of volatile and transient phenomena is paramount, attention to well-known and established features/annotations can bring important gains for text analysis.

Antal van den Bosch (Radboud University)
Monitoring stance towards vaccination in Twitter messages

As part of a collaboration between the Netherlands National Institute for Public Health and the Environment (RIVM) and the Centre for Language and Speech Technology at Radboud University, Nijmegen, we developed a system for automatically predicting the stance towards vaccination from Twitter messages, with a focus on messages with a negative stance. Such a system facilitates monitoring the ongoing stream of messages on social media, enabling swift insight into potential public discontent towards vaccination. We collected Dutch Twitter messages that mention vaccination-related key terms, and annotated them for relevance to the topic, stance towards vaccination, and sentiment. We trained machine learning classifiers to distinguish messages with a negative stance from differently categorized messages. The outcomes indicate that stance prediction is a challenging task to be dealt with by an automated system only. We see the potential of a practical setting in which a human-in-the-loop feeds the system with feedback on its predictions.


June 19th, 2018
12:00 am


VU Agora 2


Vrije Universiteit