CORA
Cork Open Research Archive (CORA) is UCC’s Open Access institutional repository which enables UCC researchers to make their research outputs freely available and accessible.
UCC Research Communities
Recent Submissions
Bacteriophage in Granular Waves
(xCoAx Conference on Computation, Communication, Aesthetics & X, 2026-07-09) Roddy, Stephen; Verdicchio, Mario; Ribas, Luísa; Rangel, André; Carvalhais, Miguel
Bacteriophage in Granular Waves is a musical work and performance that adopts data-driven composition methods, generative systems and sonification techniques to produce a musical work from a synthetic virology dataset. It has emerged from a larger collaboration exploring creative and artistic strategies for the visualization and sonification of virology data. The system employs agent-based modelling to draw samples from the dataset and uses waveshaping and granular synthesis methods to map these data to musical parameters. The end results is a performance that is always in flux where simple waveforms are sliced into tiny sections and reconfigured into texturally rich tapestries that interweave through one another shifting and unfurling in response to the virological parameters represented in the data.
Learner perceptions of study abroad in China: facilitators of L2 Chinese language learning and the role of individual attributes
(2026-12) Chen, Junming; Guo, Rongrong; Social Science Fund Project of Fujian Province; Department of Education of Fujian Province
While the number of Study Abroad (SA) students in China has increased, research on learners’ perceptions of SA experiences in China has been underexplored. Therefore, this study examines L2 Chinese learners’ perceptions of language learning facilitators during SA in China and the impacts of individual attributes (gender, proficiency, cultural background) on perceptions. A validated Likert-scale questionnaire was administered to 73 L2 Chinese learners from 22 countries, with factor analysis, nonparametric tests, and one-way ANOVAs used for data analysis. Results identified three core factors (explaining 54.937% of total variance): effective activities, supportive individuals, and key elements. Learners held more positive perceptions of the latter two more than activities. Inter-group difference analysis revealed that female learners exhibited greater recognition of most learning activities, high-level learners attached greater importance to learning opportunities and class instruction, and learners with Asian cultural backgrounds had higher recognition of language buddies. Overall, the findings demonstrate a target-oriented feature of learners’ perceptions, emphasizing in-class instruction and instructed out-of-class activities, while questioning the role of cultural events. The results call for tailored adjustments for different learners in SA programs and underscore the impact of the target language and culture on L2 learners’ perceptions. These findings provide empirical insights into the beneficial factors for language learning during SA, pointing to directions for future research.
Comparative genomics sheds new light on the microbiology of the whiskey fungus Baudoinia
(2026-08-05) Perez-Llano, Yordanis; Jackson, Stephen A.; Batista Garcia, Ramón Alberto; Dobson, Alan; European Regional Development Fund; FSE+; Horizon 2020 Framework Programme; National Council of Science and Technology (CONACyT), Government of Mexico
Baudoinia compniacensis is a melanized saprobic fungus that causes blackening of surfaces in the vicinity of spirit maturation warehouses, commercial bakeries, or distilleries that are exposed to low levels of ethanol vapour; with the fungus growing on the ethanol in the vapour. The surfaces that are most affected are those that are highly exposed and undergo extreme diurnal temperature fluctuations. This review focuses on the currently available physiological and genomic data to facilitate a comparative analysis of Baudoinia compniacensis with the genomes of a panel of fungi representing a broad spectrum of lifestyles, stress-tolerance strategies, ecological specializations and convergent adaptations that parallel those of Baudoinia, together with fungi with ethanol tolerance. By anchoring Baudoinia within this rationally selected comparative framework, this review attempts to differentiate traits that are widely shared among melanized and extremotolerant fungi from those that are unique in helping to enable the fungus colonize ethanol-rich, anthropogenic surfaces. The review also provides insights into a number of potential adaptive genetic traits that are likely to underpin the ecological success of Baudoinia, including genes involved in carbon and pentose phosphate metabolism, carbohydrate-active enzymes, genes involved in peroxisome and mitochondrial fatty-acid metabolism, together with calcium signalling and in melanin biosynthesis. We place these traits in comparative perspective, noting that many are shared across the broader oligotrophic, melanized black-fungal guild and that ethanol appears to act as a multifunctional, concentration-dependent input — germination cue, carbon source and stressor — rather than as a uniquely defining resource, with wood- and plant-derived carbohydrates likely to supplement ethanol as carbon sources. We also identify priority avenues for future work, including the sequencing of confirmed B. compniacensis, systematic survey of natural (non-industrial) reservoirs, and experimental separation of the signalling and nutritional roles of ethanol, together with direct characterisation of carbon metabolism on lignocellulosic (wood-cask) substrates.
The feature understandability scale for human-centred explainable AI: assessing tabular feature importance
(arXiv, 2025-10-10) Rossberg, Nicola; Kleinberg, Bennett; O'Sullivan, Barry; Longo, Luca; Visentin, Andrea; Taighde Éireann - Research Ireland; European Regional Development Fund
As artificial intelligence becomes increasingly pervasive and powerful, the ability to audit AI-based systems is growing in importance. However, explainability for artificial intelligence systems is not a one-size-fits-all solution; different target audiences have varying requirements and expectations for explanations. While various approaches to explainability have been proposed, most explainable artificial intelligence methods for tabular data focus on explaining the outputs of supervised machine learning models using the input features. However, a user's ability to understand an explanation depends on their understanding of such features. Therefore, it is in the best interest of the system designer to try to pre-select understandable features for producing a global explanation of an ML model. Unfortunately, no measure currently exists to assess the degree to which a user understands a given input feature. This work introduces two psychometrically validated scales that quantitatively seek to assess users' understanding of tabular input features for supervised classification problems. Specifically, these scales, one for numerical and one for categorical data, each with two factors and comprising 8 and 9 items, aim to assign a score to each input feature, effectively producing a rank, and allowing for the quantification of feature prioritisation. A confirmatory factor analysis demonstrates a strong relationship between such items and a good fit of the two-factor structure for each scale. This research presents a novel method for assessing understanding and outlines potential applications in the domain of explainable artificial intelligence.
Role of scandium precursor decomposition in MOCVD of AlScN and predictions for AlYN
(2026-07-08) Streicher, Isabel; Muriqi, Arbresha; Mullins, Rita; Straňák, Patrik; Prescher, Mario; Kapitein, Manuel; Kirste, Lutz; Quay, Rüdiger; Nolan, Michael; Leone, Stefano; Bundesministerium für Forschung, Technologie und Raumfahrt
Wurtzite Al1−xScxN is a novel nitride that attracts attention for its piezoelectric and ferroelectric properties, and further offers the possibility to grow strain-free AlScN/GaN heterostructures due to an a lattice parameter matching with GaN. Growth of AlScN by metal–organic chemical vapor deposition is challenging because of the low vapor pressure of commercially available Sc precursors and a high amount of carbon atoms present in these precursors, that can incorporate into the films. In this work, precursor decomposition and ligand loss of the novel Sc precursors (EtCp)2Sc(bdma) and (EtCp)2Sc(dtbt) are studied by density functional theory and related to experimental results. It is found that low temperatures, low pressures, and low NH3 flows promote carbon incorporation and lead to grey coloration of AlScN films grown with (EtCp)2Sc(bdma), probably due to the high residual mass of this precursor. At 1100 °C, high quality 2DEGs can be grown, especially with (EtCp)2Sc(dtbt). Predictions for the growth of AlYN/GaN heterostructures are made.