Digital Commons Network^™

Beyond The Sayable: Wittgenstein, Theory Of Mind And Affective Simulation Of Llms, Haomiaomiao Wang, Lili Zhang, Tomás E. Ward

Women+ in Early Career Research Symposium

Wittgenstein’s distinction between what can be said and what can only be shown frames a limit of propositional language. Affect and aesthetic are not primarily matters of correct description but of how expressions function within shared forms of life. Large language models (LLMs), however, increasingly produce fluent language that resembles such understanding by reproducing the patterns through which people ordinarily talk about emotion and perspective.

This paper argues that apparent Theory of Mind (ToM) in LLMs is understood as competence in the publicly shared patterns of mental-state language, rather than as grounded understanding. Using abstract artworks, we show that LLMs …

Translating Low-Level Features Into Student Friendly Explanations Using Xai, Ben Johnson, Keith Nolan

Research Outputs: 2025-Present

An alarming decline in Gaeilge proficiency has renewed education efforts, fueled by the rise of AI in recent years. This research presents the development of a deep learning framework designed to evaluate spoken Gaeilge (henceforth Irish) proficiency of learners by quantifying student voice sample similarity to everyday Irish speaker truth samples, combined with explainable AI (XAI) techniques used to generate proof-of-concept student-friendly feedback. While automatic speech recognition (ASR) systems for high-resource languages like English have access to large, detailed datasets, the Irish language relies on a limited availability of speech resources, that challenges traditional model development. We address these gaps …

Towards A Framework For Bias Evaluation Of Ai Agents In Agentic Workflows, Gary Condon, Musfira Jilani

Research Outputs: 2025-Present

Multi-agent AI systems, with minimal human oversight, are increasingly deployed in high-stakes domains like hiring, healthcare, and criminal justice. However, existing bias evaluation methodologies focus on isolated Large Language Model responses, failing to address sequential agent interactions where biases can propagate through decision chains undetected. Current approaches suffer from three limitations: absence of multi-stage bias tracking, inability to distinguish systematic discrimination from model non-determinism, and risk of contamination when agents infer bias evaluation intent. This study presents a novel evaluation framework incorporating demographic swapping methodology, contamination prevention architecture, and statistical analysis for multi-agent workflows. The framework employs dual presentation systems …

On The Automatic Image Captioning Task In Italian: A Human-Centric Approach, Valentina De Amicis, Rajesh Jaiswal, Fernando Perez-Tellez

Research Outputs: 2025-Present

Automatic Image Captioning (AIC) combines two distinct machine learning disciplines, Computer Vision (CV) and Natural Language Processing (NLP), forming a challenging research landscape. Much research has been carried out for AIC in the English language facilitated by the availability of English language datasets. However, this is not true for other languages. The aim of this work is to investigate existing AIC models' compatibility for the Italian language. The scope of this research will be aimed to reduced complexity, improved efficiency, and readily trainable models on smaller datasets while maintaining an acceptable level of caption quality. The popular Show and Tell …

The Role Of Spatial Abilities In Stem Learning And The Influence Of Individual Differences, Styliani Malkogeorgou

Masters

Students’ decisions to pursue education and careers in Science, Technology, Engineering, and Mathematics (STEM) are shaped by an interplay of cognitive, social, and motivational factors. Spatial ability is among the most reliable predictors of STEM success, yet less is known about how it relates to students’ STEM attitudes and aspirations, and whether visuospatial working memory (VSWM) explains this relationship. This study tested the hypotheses that (a) stronger spatial abilities and VSWM would be associated with more positive STEM attitudes and stronger STEM aspirations, and (b) VSWM would mediate the relationship between spatial abilities and STEM attitudes/aspirations, while examining the influence …

Segmenting Religious Tourists Based On Google Maps Reviews Using Latent Semantic Analysis And Hierarchical Clustering, Senthilkumaran Piramanayagam, Jyothi Mallya, Kartikeya Bolar

International Journal of Religious Tourism and Pilgrimage

Researchers in the tourism industry and academics are interested in analysing and extracting conclusions from user-generated data because they offer valuable insights into visitors’ motives, experiences, and consumption behaviour. However, religious tourists’ data have rarely been explored specifically in segmentation studies. Therefore, this study aims to segment religious tourists based on their Google Map reviews using Latent Semantic Analysis (LSA) and hierarchical clustering. These analyses have resulted in three distinct segments: heritage explorers, aesthetes, and self-transcendence seekers. While heritage explorers are interested in learning about the cultural and historical significance of a site, aesthetes appreciate its beauty and art, while …

Who You Gonna Call? Talk Radio As Therapy And As A Marker Of Social Change In Ireland, Finola Doyle O'Neill

Irish Communication Review

In the Irish context, the radio talk show has long constituted a distinctive cultural institution, evolving from early programmes such as Dear Frankie, The Gay Byrne Show, The Ryan Line, and Liveline into a significant public forum for the expression of personal narratives, collective opinion, and emotional resonance (O’Sullivan, 2005). Over time, this genre has undergone a notable transformation—from functioning as a secular analogue to the Catholic confessional to becoming an increasingly therapeutic outlet for listeners. This evolution is emblematic of wider sociocultural shifts in Ireland, particularly the waning influence of institutional religion (Inglis, 2000) and the disintegration of traditional …

The Road To Winning This Battle: A Corpus Informed Exploration Of Metaphor In The National Addresses Of Leo Varadkar And Micheál Martin During The Covid-19 Pandemic (2020–2022), Caoimhe Power Gallagher

Irish Journal of Applied Social Studies

The first case of the novel Covid-19 virus reached the island of Ireland in February 2020. The nation-focused response was one of the most striking characteristics that most countries shared throughout the pandemic. Corpus-linguistic research on Covid-19 communication is reasonably well established across various contexts. However, it is under researched in Ireland. Building trust and a successful national response is very much dependent on how heads of government confront health emergencies through public communications. Political leaders across the world often use metaphors in their speeches to strategically control their audiences’ perceptions of various topics. Throughout the Covid-19 pandemic, Deputy Leo …

Shap-Rc: A Framework For Explaining Annotator Disagreement In Sexism Detection, Madhuri Sawant, Arjumand Younus, Simon Caton, M. Atif Qureshi

Research Outputs: 2025-Present

The effectiveness of supervised machine learning models is heavily influenced by the quality of training data, which is often shaped by human annotators. Subjective NLP tasks such as hate speech detection, toxicity identification, and sexism classification frequently exhibit annotator disagreement due to differences in individual perspectives. This study investigates annotator disagreement in sexism detection using English tweets from the EXIST 2023 competition. To systematically analyse disagreement, tweets are categorised based on annotator consensus levels, examining how annotator demographics and linguistic features contribute to labelling inconsistencies. We interpret disagreement patterns using Shapley Additive Explanations (SHAP) and assess the consistency of SHAP-derived …

Explainable Ai For Hate Speech Moderation: A Stakeholder-Centered And Sociotechnical Review, Muhammad Deedahwar Mazhar Qureshi, M. Atif Qureshi, Wael Rashwan

Research Outputs: 2025-Present

The growing use of automated hate speech detection systems has sparked significant concern across technical, ethical, and sociopolitical dimensions. While recent advances in natural language processing and machine learning have improved classification performance, they often do so at the expense of transparency, fairness, and user trust. Explainable artificial intelligence (XAI) offers promising tools to bridge this gap, but current methods remain fragmented in scope and limited in stakeholder alignment. This review synthesizes the state of the art in explainability for hate speech detection by integrating machine learning, human-computer interaction, and critical social science perspectives. We survey major XAI approaches, including …

Techmate:: Translating Gender Equality Research Into Actionable Practice For Computing Educators, Alina Berry, Susan Mckeever, Brenda Murphy, Sarah Jane Delany

Research Outputs: 2025-Present

The persistent underrepresentation of women in computing higher education remains a global challenge, with women comprising less than 20% of ICT specialists across the EU. While substantial research has identified effective interventions to support underrepresented groups, this evidence remains fragmented across academic publications, grey literature, and institutional reports. Educators seeking to implement gender-inclusive practices face significant barriers in identifying relevant initiatives, understanding implementation requirements, and accessing practical guidance suitable for their institutional contexts. This work presents TechMate, a fully operational (https://ascnet.ie/techmate/) web-based, open-access toolkit designed to bridge the gap between research evidence and practical implementation. The toolkit consolidates over 25 …

Holding The Past Carefully In Players’ Hands: Historically-Structured Boardgames And The Troubles In History Education, Hugh D. O'Donnell

Irish Journal of Applied Social Studies

Controversial histories, such as the Troubles in Northern Ireland, are often excluded from formal education, forming a "null curriculum" due to their political sensitivity, emotional difficulty, and risk of bias or re-traumatising students. To address this, Historically-Structured Boardgames (HSBGs) offer an innovative pedagogical approach, engaging students cognitively and emotionally while navigating sensitive historical terrain.

HSBGs like "The Troubles: Shadow War in Northern Ireland 1964-1998" embed players within historically accurate socio-political and economic structures, emphasising player agency and the consequences of their actions. Players assume roles from six factions, confronting ethically charged "Narrative Cards" representing critical historical decision points. The game's …

Fuzzy Logic-Based Driving Style Classification For Lane-Change Prediction In Intelligent Transportation Systems, Muhammed Fatih Koc, Nouman Ashraf, Pramod Pathak, Sachin Sharma

Research Outputs: 2025-Present

In recent years, Intelligent Transportation Systems (ITSs) have emerged as a solution to mitigate the problem of traffic congestion. Understanding human driving styles such as aggressive, normal, and cautious is crucial for safe driving. In particular, predicting lane-change manoeuvres may be further supported by combining vehicle state information with driving style information. However, existing vehicle trajectory datasets lack driving style information, making classification challenging. To address this limitation, this paper proposes a fuzzy logic-based driving style classification framework in a Vehicle-to-Everything (V2X) environment. The model uses vehicle state information, including speed, longitudinal acceleration, lateral acceleration, and distance headway to classify …

Educating The Next Generation Of Engineers: Spatial Thinking Is Key To Their Success, Sheryl Sorby, Gavin Duffy, Jack Parkinson, Gibin Raju, John Lynch, Muhammad Asghar

Research Outputs: 2025-Present

This paper attempts to answer questions regarding the link between engineering success and spatial skills through three themes. The previous research of the authors in examining the questions surrounding each theme is the focus of this paper; we acknowledge that others around the world may have conducted similar research, but the specific findings of the authors are the main focus of this paper. In the first theme, we examine data from several author-led studies where positive correlations have been found between spatial skill levels and proficiency in tasks that are a part of what it means to be an engineer. …

Diffusion-Enhanced Nat-Bart Vision Language Transformer For Unified Urdu Word Recognition, Wahid Hussain, Shahbaz Hassan, Isma Ul Hassan, Ahmad Raza Shahid, Asif Naeem, Muhammad Usman Yaseen, Nouman Ashraf

Research Outputs: 2025-Present

Handwritten Urdu text recognition remains a very challenging problem due to the cursive and complex nature of the Nastaliq script. Each writer has a unique style, and there is a severe lack of large, well-labeled datasets. These challenges make Urdu text recognition much harder than for many other languages. Here we propose a novel and effective approach for handwritten Urdu word recognition that addresses both data scarcity and feature complexity. We utilized conditional diffusion models to generate synthetic handwritten Urdu word images. These models produce realistic word samples covering a wide range of variations, such as stroke thickness, curves, and …

Historic Building Information Modelling - Adding Intelligence To Laser And Image Based Surveys, Maurice Murphy Dr, Eugene Mcgovern, Sara Pavia

Journal of Academic Research and Innovation

Historic Building Information Modelling (HBIM) is a novel prototype library of parametric objects based on historic data and a system of cross platform programmes for mapping parametric objects onto a point cloud and image survey data. The HBIM process be-gins with remote collection of survey data using a terrestrial laser scanner combined with digital photo modelling. The next stage in-volves the design and construction of a parametric library of objects, which are based on the manuscripts ranging from Vitruvius to 18th century architectural pattern books. In building parametric objects, the problem of file format and exchange of data has been …

Relational Pedagogy, Student Led Seminars, And Rethinking The World ‘Otherwise’: Evidence From Focus Groups With Ecce Students In An English University, Lisa Moran, Elizabeth Richards Ms

Irish Journal of Applied Social Studies

This paper critically examines findings from a qualitative study of collaborative, student-led seminars framed within relational pedagogy, with undergraduate students on an Early Childhood Care and Education (ECCE) degree programme in England. Drawing upon focus groups and observational data, this paper argues for the importance of student-led seminars in ECCE programmes, for improved student’s motivation to learn, teamwork, collaboration skills and peer learning. Significantly, our data reveals that depth engagement in these seminars facilitated students rethinking their worlds ‘otherwise’ where they questioned their own biases and prejudices towards so-called hard to reach and marginalised groups. Utilising a relational pedagogical approach …

Transdisciplinary Perspectives On Ai: The Fourth Annual Conference Of The European Culture And Technology Laboratory, Connell Vaughan, Ioana Madalina Moldovan, Silivan Moldovan, Noel Fitzpatrick

Books/Book Chapters

The fourth annual conference of the ECT Lab+ was hosted by Technical University of Cluj-Napoca over two days in October 2024 at the Cluj Innovation Park. The conference brought together experts from the Arts, Humanities, Social Sciences, Technology, and other fields to discuss and reflect on the advent of Artificial Intelligence and how the associated technologies are transforming how we live, work and study. Under the title Transdisciplinary perspectives on AI: Alternative Histories, Current Practices and Possible Futures the conference moved beyond simplistic technophila and technophobia to consider whether we can co-evolve with these new technologies which combine machine learning …

The History Of Dublin Coddle And Its Significance As Part Of Ireland’S Intangible Cultural Heritage, Sinéad Mcglynn, Máirtín Mac Con Iomaire

European Journal of Food Drink and Society

Dublin coddle is held in great affection by generations of Dubliners and is a significant cultural identifier, yet to date there has been little or no academic research into its historical origins. Dublin coddle is a divisive dish which appears to be linked with the capital city, with almost tribal associations, where other ‘tribes’ refuse to consume it. This study argues for the inclusion of Dublin coddle-making, a regional dish made with sausages, ham or bacon, potatoes and onions, as part of Ireland’s Intangible Cultural Heritage (ICH) as set out by the United Nations Educational, Scientific and Cultural Organization (UNESCO). …

Food Traditions In The Folklore Of Oíche Shamhna (Halloween): An Exploratory Study Of Food In The Schools’ Collection 1937–39, Caitríona Nic Philibín

European Journal of Food Drink and Society

Oíche Shamhna or Halloween marks the end of the harvest, the beginning of winter and the last of the four Quarter Days of the Irish calendar year. While food is plentiful at this time of year, food specific studies of this feast day are few. Through archival research this paper explores food traditions at Oíche Shamhna found in the Schools’ Collection 1937–’39 (folklore collected from more than 50,000 schoolchildren in Ireland). Analysis of a sample taken from the Collection reveals four main themes: ‘Games,’ ‘Games with Divinations,’ ‘Supernatural,’ and ‘Cakes and dishes in general.’ Within these themes the significance of …

Gerunds In Irish Sign Language: An Exploratory Analysis, Zaid Mohammed

Doctoral

In sign languages (SLs), e.g., American Sign Language (ASL), gerunds help express the meaning of ongoing actions and activities conveyed by verbs (Klima & Bellugi, 1979, p. 295). In spoken languages, a gerund is a form of the verb that functions as a noun and expresses the state of being or an ongoing action (Maekelberghe, 2020). Gerunds are complex constructions as they have nominal-verbal nature (Travis, 2005, p. 321) and morphological similarity with present participle or progressive aspect (Siegel, 1998). This research is motivated by the challenging nature of distinguishing gerund constructions at the morphological level, and research in sign …

Communicating The Green Transition - Practices And Competencies For Engineers, Paul Talbot, Una Beagon, Stephen Harper, Janet Mckennedy

Research Outputs: 2025-Present

The green transition is driving demand for engineering expertise in sectors such as renewable energy, construction, transport, and industrial decarbonisation. Projects are complex and interdisciplinary, and communication plays a central role. This report presents findings from 20 focus groups with 103 engineers across Europe examining how engineers communicate in green transition project contexts. Participants described a wide range of communicative tasks involving coordination, documentation, stakeholder engagement, and public communication.Engineers work across spoken, written, and digital modes, shifting between formal and informal styles, and adapting to varied audiences. They must navigate fragmented tools and workflows, unclear goals and roles, jargon pitfalls, …

Image Captioning Through The Lens Of The Gricean Maxims: Generating Meaningful And Relevant Image Descriptions, Annika Lindh

Doctoral

Image captioning models enable us to automatically generate natural language image descriptions for previously unseen images. It combines the two fields of computer vision and natural language generation, allowing models to interpret the con tent of an image and communicate that knowledge through natural language text.

Research into image captioning has the potential benefit of reducing the gap in digital information availability between fully sighted individuals and those who are visually impaired. However, automatically generated captions often fail to provide the required level of detail and specificity to achieve this goal. Furthermore, current standard evaluation methods are insufficient at measuring …

Fostering Sustainability Literacy And Action Through Language Education: Perspectives And Practices Across Regions, Maria-José De La Fuente, Odette Gabaudan, Kiley Kost, Owain Llewellyn, Pilar Molina, Seth Peabody

Research Outputs: 2025-Present

This activity report explores the integration of Education for Sustainable Development (ESD) in language pedagogy, focusing on its potential to equip students with both linguistic proficiency and the competencies needed to address complex global challenges. Rooted in the context of TU (Technical University) Dublin’s commitment to embedding sustainability in Higher Education curricula, the activity report builds on insights from a 2023 roundtable discussion convened by two of the authors in collaboration with CercleS and the United Nations Institute for Training and Research (UNITAR) on the occasion of its first Research conference. Featuring perspectives from scholars in sustainability and language education, …

Lomatce: Local Model-Agnostic Time Series Classification Explanations, Ephrem Tibebe Mekonnen, Luca Longo, Pierpaolo Dondio

Conference papers

Deep learning models perform exceptionally well in time series classification, but their lack of interpretability remains a significant challenge. Although explainable AI (XAI) techniques are well established for image and tabular data, adapting these methods to time series is challenging due to temporal dependencies. Existing feature importance methods often treat each time step as an independent feature, ignoring temporal relationships or segmenting the series into fixed-size windows, which can lead to the omission of critical subsequences. This paper introduces a novel, local, model-agnostic interpretability method designed explicitly for time series data without segmentation. The technique generates neighbouring samples by randomly …

Sustainfed-Llm: Renewable Energy Aware Client Selection For Energy Efficient Federated Training Of Large Language Models, Sunbal Iftikhar, Hassan Khan, Steven Davy

Research Outputs: 2025-Present

Federated Learning (FL) offers a decentralized approach to training large language models (LLMs), addressing critical concerns around data privacy and transmission costs. However, FL's inherent distributed training paradigm can lead to increased energy consumption and carbon emissions, especially with random or uninformed client selection. To tackle this issue, we propose SustainFed-LLM, a novel Q-learning-based client selection framework that integrates real-time renewable energy availability, carbon intensity data, and fairness considerations. By dynamically assessing client performance, spare capacity, and sustainability metrics, SustainFed-LLM optimizes client participation to minimize environmental impact while maintaining model accuracy. SustainFed-LLM significantly reduces energy consumption by up to 50%, …

Codfish In Portugal: Drafting A Proposition Of A National Tourist Itinerary, Cristina Carvalho, António Gonçalves

International Journal of Tour Guiding and Visitor Attraction Research

The purpose of this empirical study is to draft a proposition on a tourist itinerary feasible of being organised, promoted and interpreted in its technical features by stakeholders, namely tour operators, promoters and official guides across Portugal. This themed proposition on codfish links towns such as Viana do Castelo, Ílhavo and Lisbon, thus encompassing manifold cultural expression elements, from cod schooners to museums, and from gastronomy to events. As an empirical study, the authors resort to data collected from several scientific sources to base its (yet brief) literary review on cultural tourism and heritage interpretation. Their experience in tour guiding …

From Affordances To Action: Designing Familiar Health Data Interfaces For Older Adults, Peterson Jean, Enda Bates, Emma Murphy Dr

Conference papers

Health data is inherently complex, and tracking it appropriately, whether prescribed by a doctor or motivated by personal interest, offers significant benefits for improving health behaviours. With consumer availability, pervasive technologies such as wearable devices have opened new opportunities for the self-tracking of critical physiological health data. Yet, despite the progress made, older adults often encounter barriers that limit meaningful interactions, leading them to disengage from managing their own health. At the root of these barriers is the gap between designers' intentions and older adults' ability to perceive, interpret, and respond effectively to cues embedded in systems designed to collect …

Evaluation Of Pre-Hajj Multimedia Training Tools: A Case Study Of Prospective Australian Pilgrims, Kholoud M. Munshi, Mustapha Taibi Dr., Kate Richards Ms., Khushnoor Khan Mr.

International Journal of Religious Tourism and Pilgrimage

Every year, millions of Muslim pilgrims gather in Makkah (Saudi Arabia) for Hajj. These pilgrims come from different parts of the world, with diverse language and cultural backgrounds. Managing such a large and diverse congregation for a successful completion of Hajj requires effective planning, which depends on two important antecedents—knowing the background information of the pilgrims and developing effective pre-Hajj training tools for dissemination of Hajj information for non-Arabic speaking pilgrims. This study evaluates the perceptions of potential Hajj pilgrims regarding community translation usability (CTX) and medium usability (MX) of two existing pre-Hajj media tools developed by Mnask Academy (Booklet …

The Complexity Of Long-Distance Dependencies And Their Impact On Language Models, Abhijit Shrikant Mahalunkar

Doctoral

Sequential data modeling is an important challenge in various fields and in particular in natural language processing. Building effective sequential models faces a notable challenge in the form of Long-Distance Dependencies (LDDs) within the sequence data. Hence, successfully overcoming this challenge is imperative for developing robust and accurate sequential models across various domains and applications. To tackle this challenge, the first step is to conduct a detailed analysis of the complexity of LDDs observed in various sequence datasets. This thesis offers a thorough exploration and documentation of this analysis. An important finding from this thesis is the consistent patterns of …