Chris Emmery

Chris Emmery

Auto-generated profile, based on publicly available information

Chris Emmery

Tilburg University

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 October 2018 in PLOS ONE
Verified
Journal Article
Open access
PDF file
CHCynthia Van Hee
Cynthia Van Hee
Gilles M. JacobsGilles M. Jacobs
Gilles M. Jacobs
CEChris Emmery
Chris Emmery
BDBart Desmet
Bart Desmet
ELEls Lefever
Els Lefever
VHVéronique Hoste
Véronique Hoste
While social media offer great communication opportunities, they also increase the vulnerability of young people to threatening situations online. Recent studies report that cyberbullying constitutes a growing problem among youngsters. Successful prevention depends on the adequate detection of potentially harmful messages and the information overload on the Web requires intelligent systems to identify potential risks automatically. The focus of this paper is on automatic cyberbullying detection in social media text by modelling posts written by bullies, victims, and bystanders of online bullying. We describe the collection and fine-grained annotation of a cyberbullying corpus for English and Dutch and perform a series of binary classification experiments to determine the feasibility of automatic cyberbullying detection. We make use of linear support vector machines exploiting a rich feature set and investigate which information sources contribute the most for the task. Experiments on a hold-out test set reveal promising results for the detection of cyberbullying-related posts. After optimisation of the hyperparameters, the classifier yields an F1 score of 64% and 61% for English and Dutch respectively, and considerably outperforms baseline systems.

16 November 2020 in Language Resources and Evaluation
Verified
Journal Article
Open access
PDF file
CEChris Emmery
Chris Emmery
BVBen Verhoeven
Ben Verhoeven
GPGuy De Pauw
Guy De Pauw
Gilles M. JacobsGilles M. Jacobs
Gilles M. Jacobs
CHCynthia Van Hee
Cynthia Van Hee
WDWalter Daelemans
Walter Daelemans
The detection of online cyberbullying has seen an increase in societal importance, popularity in research, and available open data. Nevertheless, while computational power and affordability of resources continue to increase, the access restrictions on high-quality data limit the applicability of state-of-the-art techniques. Consequently, much of the recent research uses small, heterogeneous datasets, without a thorough evaluation of applicability. In this paper, we further illustrate these issues, as we (i) evaluate many publicly available resources for this task and demonstrate difficulties with data collection. These predominantly yield small datasets that fail to capture the required complex social dynamics and impede direct comparison of progress. We (ii) conduct an extensive set of experiments that indicate a general lack of cross-domain generalization of classifiers trained on these sources, and openly provide this framework to replicate and extend our evaluation criteria. Finally, we (iii) present an effective crowdsourcing method: simulating real-life bullying scenarios in a lab setting generates plausible data that can be effectively used to enrich real data. This largely circumvents the restrictions on data that can be collected, and increases classifier performance. We believe these contributions can aid in improving the empirical practices of future research in the field.

12 May 2025 in PLOS ONE
Verified
Journal Article
Open access
PDF file
RRRonja Rönnback
Ronja Rönnback
CEChris Emmery
Chris Emmery
HBHenry Brighton
Henry Brighton
Political bias is an inescapable characteristic in news and media reporting, and understanding what political biases people are exposed to when interacting with online news is of crucial import. However, quantifying political bias is problematic. To systematically study the political biases of online news, much of previous research has used human-labelled databases. Yet, these databases tend to be costly, and cover only a few thousand instances at most. Additionally, despite the wide recognition that bias can be expressed in a multitude of ways, many have only examined narrow expressions of bias. For example, most have focused on biased wording in news articles, but ignore bias expressed when an outlet avoids reporting on certain topics or events. In this article, we introduce a data-driven approach that uses machine learning techniques to analyse multiple forms of bias, and that can estimate the political leaning of hundreds of thousands of Web domains with high accuracy. Crucially, this approach also allows us to provide detailed explanations for why a news outlet is assigned a particular political bias. Our work thereby presents a scalable and comprehensive approach to studying political bias in news on a larger scale than ever before.

14 August 2019 in Computational Brain & Behavior
Verified
Journal Article
Open access
PDF file
CEChris Emmery
Chris Emmery
ÁKÁkos Kádár
Ákos Kádár
TWTravis J. Wiltshire
Travis J. Wiltshire
AHAndrew T. Hendrickson
Andrew T. Hendrickson
The suggestions proposed by Lee et al. to improve cognitive modeling practices have significant parallels to the current best practices for improving reproducibility in the field of Machine Learning. In the current commentary on “robust modeling in cognitive science”, we highlight the practices that overlap and discuss how similar proposals have produced novel ongoing challenges, including cultural change towards open science, the scalability and interpretability of required practices, and the downstream effects of having robust practices that are fully transparent. Through this, we hope to inform future practices in computational modeling work with a broader scope.

29 July 2019 in PsyArXiv
Verified
Preprint
Open access
CEChris Emmery
Chris Emmery
ÁKÁkos Kádár
Ákos Kádár
TWTravis J. Wiltshire
Travis J. Wiltshire
AHAndrew T Hendrickson
Andrew T Hendrickson
The suggestions proposed by Lee et al. to improve cognitive modeling practices have significant parallels to the current best practices for improving reproducibility in the field of Machine Learning. In the current commentary on `Robust modeling in cognitive science', we highlight the practices that overlap and discuss how similar proposals have produced novel ongoing challenges, including cultural change towards open science, the scalability and interpretability of required practices, and the downstream effects of having robust practices that are fully transparent. Through this, we hope to inform future practices in computational modeling work with a broader scope.

25 October 2019 in arXiv
Verified
Preprint
Open access
PDF file
CEChris Emmery
Chris Emmery
BVBen Verhoeven
Ben Verhoeven
GPGuy De Pauw
Guy De Pauw
GJGilles Jacobs
Gilles Jacobs
CHCynthia Van Hee
Cynthia Van Hee
WDWalter Daelemans
Walter Daelemans
The detection of online cyberbullying has seen an increase in societal importance, popularity in research, and available open data. Nevertheless, while computational power and affordability of resources continue to increase, the access restrictions on high-quality data limit the applicability of state-of-the-art techniques. Consequently, much of the recent research uses small, heterogeneous datasets, without a thorough evaluation of applicability. In this paper, we further illustrate these issues, as we (i) evaluate many publicly available resources for this task and demonstrate difficulties with data collection. These predominantly yield small datasets that fail to capture the required complex social dynamics and impede direct comparison of progress. We (ii) conduct an extensive set of experiments that indicate a general lack of cross-domain generalization of classifiers trained on these sources, and openly provide this framework to replicate and extend our evaluation criteria. Finally, we (iii) present an effective crowdsourcing method: simulating real-life bullying scenarios in a lab setting generates plausible data that can be effectively used to enrich real data. This largely circumvents the restrictions on data that can be collected, and increases classifier performance. We believe these contributions can aid in improving the empirical practices of future research in the field.