SafeSpacesNLP: Exploring behaviour classification around online mental health conversations from a multi-disciplinary context - NLP, applied linguistics, social science and human-in-the-loop AI

Dr. Stuart Middleton1 & Dr Elena Nichele2

1University of Southampton, ECS  2University of Nottingham

In this seminar, we will present the UKRI TAS hub SafeSpacesNLP research project. The SafeSpacesNLP project is exploring how multi-disciplinary teams can use NLP-based behaviour classification algorithms to identify online harmful behaviours for children and young people. We will present our research from the viewpoint of Computer Science and the viewpoint of Applied Linguistics / Social Science, highlighting both the existing research challenges for each discipline and how SafeSpacesNLP is trying to address some of these. We will conclude with the longer-term challenges we see for this research area.

Dr Stuart Middleton:
Dr Stuart E. Middleton is an Associate Professor at the University of Southampton, Electronics and Computer Science (ECS), AIC Group. He has over the last 17 years made internationally recognized contributions to research into natural language processing and information extraction. Many of his grants are cross-disciplinary in nature, featuring consortia with a mixture of academic and commercial partners experienced in a range of domains and disciplines.

He is currently PI of NERC-funded platform grant GloSAT (Global Surface Air Temperature - NLP for data rescue) and UKRI TAS Hub funded SafeSpacesNLP (Behaviour classification NLP for online harmful behaviours for children and young people). He is CoI of ESRC-funded grant FloraGuard (Tackling the illegal trade in endangered plants - Socio-technical NLP) and ESRC-funded ProTecThem (Building Awareness for Safer and Technology-Savvy Sharenting).

He is a Turing Fellow, was chair of the recent WebSci'20 Workshop 'Socio-technical AI systems for defence, cybercrime and cybersecurity' and is a member of the Sector Leads Committee of the UKRI Trustworthy and Autonomous Systems (TAS) Hub. He has been an invited AI expert at both the ATI/DSTL workshop 2019 on 'Decision Support for Military Commanders' and the UK Cabinet Office ministerial AI roundtable event 2019 on 'use of AI in policing' which was chaired by the policing minister.

Dr Elena Nichele:
Dr Elena Nichele is a Research Fellow at the School of Computer Science of the University of Nottingham (UK), where she is a member of the Horizon Digital Economy Research Institute and the Trustworthy Autonomous Systems Hub and she is involved in different multi-disciplinary projects.

Her main research interests lie in business communication, human-computer interactions, cross-cultural exchanges, and computer-mediated communication. She has been using linguistic approaches to examine perceptions surrounding social, cultural and political matters. Her previous research has involved corpus linguistics, (critical) discourse analysis, commodification, and digital marketing. In particular, she has been focusing on the concept (and communication) of (in)authenticity.

Her expertise combines linguistics and marketing. She was awarded the PhD title in Applied Linguistics from Lancaster University (UK) and holds degrees in International Business (MA) from the University of Florida (USA) and Languages for Business Communications (MA) from the University of Verona (Italy).

Week 22 2021/2022

Thursday 5th May 2022

Microsoft Teams