Talk Title: Repurposing in AI: A Distinct Approach or an Extension of Creative Problem Solving?

The talk was delivered on 26 February 2025 by Aissatou Diallo, as part of the DIS research seminars series.

Creativity is defined as the ability to produce novel, useful, and surprising ideas. A sub area of creativity is creative problem solving, the capacity of an agent to discover novel and previously unseen ways to accomplish a task, according to its perspective. While creative problem solving has been extensively studied in AI, the related concept of repurposing – identifying and utilizing existing resources in innovative ways to address different problems from their intended purpose – has received less formal attention. This paper presents a theoretical framework that distinguishes repurposing from creative problem solving by formalizing both approaches in terms of conceptual spaces, resource properties, and goal achievement mechanisms. We demonstrate that while creative problem solving involves expanding the conceptual space through transformation functions, repurposing operates within existing conceptual spaces by leveraging shared properties of available resources. This formalization provides new insights into how these two approaches to problem-solving differ in their fundamental mechanisms while potentially complementing each other in practical applications..

Talk Title: Emotion Elicitation, Emotional Experience and Recognition of Emotion

The talk was delivered on 22 January 2025 by Daniela Romano, Honorary Lecturer at the Department of Information Studies, as part of the DIS research seminars series.

This talk presents Professor Romano’s latest research findings in Digital Affective Science at the crossroads between psychology and computer science, and the qualitative, quantitative and AI methods used in this area illustrated with various research projects. The talk presents an artistic robotics installation that induces mindfulness, an investigation into the Autonomous sensory Meridian response phenomenon, emotions and behavioural tendencies triggered by extensive video game play, and emotions recognition from WiFi signals.

Professor Daniela Romano received her PhD in supporting naturalistic decision-making with an intelligent serious game in 2002 at Leeds University, Lecturer in Computer Science (2004) & Senior Lecturer (2010), & Virtual Reality Theme Leader at Sheffield. Professor in Computing (2015) & Head of Innovative Research and Enterprise, Manager of Virtual Reality at Edge Hill, Senior Teaching Fellow at UCLIC in 2017, Senior Teaching Fellow at UCL Information Studies in 2019, Professor of Artificial Intelligence at de Montfort University and Director of the Institute of Digital media, Artificial intelligence, Responsible innovation, Ethics and Cybersecurity.

Ethical Data use in Irish public policy: the role of the Data Governance Board

The talk was delivered on 11 November 2024 by Clare Thornley, Honorary Research Associate at the Department of Information Studies, as part of the DIS research seminars series.

The Data Governance Board was established in Ireland in 2021 under the Data Sharing Act 2019 and it plays a key role in developing and guiding the Public Service Data Strategy in Ireland. This seminar will introduce the role and remit of the Data Governance Board. Its focus will be on the work of its Data Safeguarding and Transparency Committee which works to ensure that any transparency, privacy or data protection concerns that arise around the sharing or governance of data are managed in an ethical way. This includes the development of a Data Sharing Ethics Framework to assist public servants in implementing ethical standards in all aspects of data use. It is an interesting example of efforts to integrate ethical values into processes and methods and this will be discussed considering wider work in information and data ethics. The role of the Data Governance Board in shaping the next Public Service Data Strategy will also be discussed examining emerging challenges of technology such as AI as well as persistent trade-offs that need to be managed.

Clare Thornley is Board Member of the Data Governance Board, Ireland.
Clare Thornley has an MA (Hons) in Philosophy from the University of Edinburgh, an MSc in Information Management and a PhD in Information Retrieval from the University of Strathclyde. She has been an Honorary Research Associate at UCL since 2011. She started her career as an Information Officer in the UK voluntary sector at The Volunteer Centre UK before completing her PhD and moving into research. She has lectured on Information Retrieval at University College Dublin and Dublin Business School and currently teaches on the online Information Systems MSc at the University of Liverpool. As well as being a Data Governance Board member since 2021 Clare works as an independent consultant on EU and international digital skills and IT capability development projects with a focus on developing ethical standards and guidelines. Her recent publications include an EU Ethics Framework for ICT Professionals and a research paper on the use of Science Fiction to inform responsible innovation. She has also served on the Ethics Advisory Board of two EU H2020 Research Projects on AI and law enforcement.

The Imitation Game: Advancing guidance on AI ethics and governance in practice

The talk was delivered on 19 March 2024 by Dr. Ann Borda, an Ethics Fellow in the Public Policy Programme at The Alan Turing Institute. and an Honorary Senior Research Associate in the Department of Information Studies, as part of the DIS research seminars series.

AI is having a significant impact on public policies and services around the world, but government use of AI has a steep learning curve, and the purpose of AI within government and public sector contexts present numerous challenges. To help UK civil servants learn about and explore AI in an effective and ethical way, the Alan Turing Institute’s Public Policy Programme developed a series of workbooks that promotes the understanding of the UK Government’s official Public Sector Guidance on AI Ethics and Safety published in 2019, in collaboration with the UK’s Office for Artificial Intelligence and the Government Digital Service. Co-developed with public sector groups, this guidance outlines how AI project teams in the public sector can put ethical values and practical principles into practice across the AI project lifecycle, ensuring that AI is produced and used ethically, safely, and responsibly. Complementary to this initiative, the Turing is growing a societal Readiness, Skills, and Knowledge platform which further includes guidance on AI ethics and skills tracks for early career researchers and community audiences, supported by a publicly accessible repository of resources for those seeking to explore and apply the ethical and responsible use of data. These initiatives, including the underlying ethical values and frameworks which underpin them, are the key focus of this seminar. Challenges of the evolving AI landscape are also touched on, particularly in the development and deployment of guidance for multiple stakeholders.

Incorporating Dictionaries into a Neural Network Architecture to Extract COVID-19 Medical Concepts From Social Media

The talk was delivered on 24 January 2024 by Dr. Abul Hasan, a postdoctoral research fellow at the UCL Institute of Health Informatics, as part of the DIS research seminars series.

We investigate the potential benefit of incorporating dictionary information into a neural network architecture for natural language processing. In particular, we make use of this architecture to extract several concepts related to COVID-19 from an on-line medical forum. We use a sample from the forum to manually curate one dictionary for each concept. In addition, we use MetaMap, which is a tool for extracting biomedical concepts, to identify a small number of semantic concepts. For a supervised concept extraction task on the forum data, our best model achieved a macro F1 score of 90%. A major difficulty in medical concept extraction is obtaining labelled data from which to build supervised models. We investigate the utility of our models to transfer to data derived from a different source in two ways. First for producing labels via weak learning and second to perform concept extraction. The dataset we use in this case comprises COVID-19 related tweets and we achieve an F1 score 81% for symptom concept extraction trained on weakly labelled data. The utility of our dictionaries is compared with a COVID-19 symptom dictionary that was constructed directly from Twitter. Further experiments that incorporate BERT and a COVID-19 version of BERTweet demonstrate that the dictionaries provide a commensurate result. Our results show that incorporating small domain dictionaries to deep learning models can improve concept extraction tasks. Moreover, models built using dictionaries generalize well and are transferable to different datasets on a similar task.

Towards Computational Persuasion for Behaviour Change Applications.

The talk was delivered on 22 November 2024 by Prof. Anthony Hunter, Professor of Artificial Intelligence in the Department of Computer Science, University College London, as part of the DIS research seminars series.

The aim of behaviour change is to help people to change aspects of their behaviour for the better (e.g., to decrease calorie intake, to drink in moderation, to take more exercise, to complete a course of antibiotics once started, etc.). Recent developments in computational modelling of argument (a subfield of AI) are leading to technology for persuasion that can potentially be harnessed in behaviour change applications. Using this technology, a software system and a user can exchange arguments in a dialogue. So the system gains information about the user’s perspective, provides arguments to fill gaps in the user’s knowledge, and attempts to overturn misconceptions held by the user. Our work has focused on modelling the beliefs and concerns of the user, and harnessing these to make the best choices of move during the dialogue for persuading the user to change their behaviour. We have also been investigating how we can harness recent developments in large language models to provide a natural language interface to this technology. In this talk, I will provide an overview of our approach together with some promising preliminary results with participants.