11th SIKS/Twente Seminar on Searching and Ranking

Seminar on Cyberbullying


The goal of this seminar is to bring together researchers from academia and organizations working on the development of strategies and solutions to understand, detect and prevent cyberbullying incidents among adolescents. Invited speakers are:

The symposium will take place at the campus of the University of Twente in building Zilverling, room 2042.
See Travel information. The event is part of the SIKS educational program. PhD-students working in the field of (interactive) information filtering, recommending, and retrieval are encouraged to participate.


Coffee and Welcome
Opening and Introduction
Prof. Franciska de Jong (University of Twente, the Netherlands)
Preventing Bullying: A Focus on Healthy Relationships
Prof. Debra Pepler (York University, Canada)

Through our observational research, we have come to understand bullying as a relationship problem in which power is used aggressively to control and distress others. Our longitudinal research highlighted different pathways of bullying from elementary to high school and the links between parent and peer relationships and youths' problems of bullying and victimization. Emerging research on genetics and neurodevelopment highlight the importance of healthy relationships for healthy development. Over the past 8 years, we have been bridging the gap between science and practice related to healthy relationships and development through a national network, PREVNet – Promoting Relationships and Eliminating Violence Network. With over 60 national partner organizations, we have been mobilizing knowledge into the hands of those who are involved with children and youth in all the places where they live, learn, and play.

Recognising Suicidal Thoughts in Social Media
Prof. Veronique Hoste (Ghent University, Belgium)

Early detection of suicidal thoughts is an important part of effective suicide prevention. Such thoughts may be expressed online, especially by young people. In this talk, we will present on-going work on the automatic recognition of suicidal messages in social media. We present experiments for automatically detecting relevant messages (with suicide-related content), and those containing suicide threats. Together with volunteers of the Belgian Suicide Centre, we collected and annotated online data based on a jointly developed fine-grained annotation scheme. The texts were annotated with information on relevance, origin, severity of suicide threat and risks as well as protective factors. A machine learning set-up was used, exploiting both shallow lexical and deeper semantic features to predict relevance and severity. We report on the experimental results and show by means of feature optimisation through the use of genetic algorithms which factors are decisive for the detection of relevant and severe suicidal messages.

Experts and Machines United Against Cyberbullying
PhD Defense (location: Waaier 4) by Maral Dadvar (University of Twente)

One from of online misbehaviour which has deeply affected society with harmful consequences is known as cyberbullying. Cyberbullying can simply be defined as an intentional act that is conducted through digital technology to hurt someone. Cyberbullying is a widely covered topic in the social sciences. However, studies on the technical dimensions of this topic are relatively rare. In this research the overall goal was to bridge the gap between social science approaches and technical solutions have thoroughly studied the origin of cyberbullying and its growth over time, as well as the role of technology in the emergence of this type of virtual behaviour and in the potential for reducing the extent of the social concern it raises.
This thesis introduces a novel outlook towards the cyberbullying phenomenon. It looks into the gradual changes which have occurred in relationships and social communication with the emergence of the Internet. This study leads to the conviction that for combating cyberbullying, behavioural and psychological studies, and the study of technical solutions should go hand in hand. It was also shown that besides the conventional features used for text mining methods more personal features, such as gender and age of the users, can improve the accuracy of the detection models. We also study the use of experts' knowledge and human reasoning to identify potential bully users in social networks. A hybrid approach is designed, incorporating machine learning models on top of the expert system, which outperforms each of the models individually. The integration of social studies into a software-enhanced monitoring workflow could pave the way towards the tackling of this kind of online misbehaviour. This work can be viewed as a contribution to the more general societal challenge of increasing the level of cybersecurity, in particular for the younger generations of social network users.


CTIT Center for Data Science
Netherlands research school for Information and Knowledge Systems


Please send your name and affiliation to if you plan to attend the symposium, and help us estimate the required catering.