13463 research outputs found
Sort by
Editorial: Empowering individuals: promoting health literacy through curriculum and science communication.
No full textEditorial on the Research Topic Empowering individuals: promoting health literacy through curriculum and science communicatio
The Behavior Specialist in Inclusive Schools: Navigating Power, Support, and Intervention for Behaviours of Concern
No full textBehaviors of concern (BoC) may be defined as persistent behaviors that impact the daily functioning and learning of children. They are behaviors that could pose a risk to their own safety or the safety of others. Supporting children with BoC is vital for student learning, success, and inclusion in both mainstream and special settings. Traditionally, the
onus has been on the teacher to support and manage a classroom and all behaviors within that classroom. However, with an increase in BoC impacting school and class activities, targeted support in schools has become more common. Many factors have accelerated this shift, particularly the rise of burnout, lack of confidence, and aggressive behavior in schools, particularly in special school settings. The current study, through a Foucauldian power/knowledge and disciplinary theory paradigm, investigated whether having one specialist in the school aids the children and staff. A focused case study was conducted via
seven semi-structured interviews with staff at one special school in the Republic of Ireland (RoI) with a full-time behavior specialist (BS) on site. The findings revealed that (1) the role of the BS is based on identifying BoC and implementing support, (2) having a behavior specialist is key for supporting children and staff in schools, (3) support and interventions
are more sustainable once there is the presence of a specialist, and (4) challenges such as a lack of space, inadequate funding, large caseloads, staff shortages, and lack of time are a reality in the school setting. The main conclusion derived from this study is that having a BS in the school has a positive impact on the children, staff, and attitudes, providing both
practical and pastoral power, which are essential for effective inclusive practices
How Emotions Can Help Detect Synthetic Text
Get PDFQuestion: Can you tell whether any of this thesis was written by AI? Recent developments in generative AI have shone a spotlight on high performance synthetic text generation technologies. The wide availability and ease of use of such models highlights the urgent need to provide equally powerful technologies capable of identifying synthetic text. With this in mind, we draw inspiration from psychological studies which suggest that people can be driven by emotion and encode emotion in the text they compose. We hypothesise that pretrained language models (PLMs) have an affective deficit because they lack such an emotional driver when generating text and consequently may generate synthetic text which has affective incoherence i.e. lacking the kind of emotional coherence present in human-authored text. We subsequently develop an emo- tionally aware detector by fine-tuning a PLM on emotion. Experiment results indicate that our emotionally-aware detector achieves improvements across a range of synthetic text generators, various sized models, datasets, and domains. We compare our emotionally-aware synthetic text detector to ChatGPT in the task of identification of its own output and show substantial gains, reinforcing the potential of emotion as a signal to identify synthetic text. These findings support the hypothesis that PLMs may have an affective deficit. Next, we investigate the hypothesis that synthetic text may be affectively incoherent. We create a novel flexible evaluation framework and use it to select an emotion classifier to generate an affective profile for 10k human and synthetic news arti- cles. Our analysis of the human and synthetic affective profiles show that they are similar, but synthetic text is more affectively incoherent and less affectively coherent, than human text. Answer: AI wrote none of this thesis, but how do you know for certain? This lack of certainty motivates the task of synthetic text detection
Investigating production routes and application of nanotechnology for improved properties of porous copper structures
Get PDFThis master thesis presents the investigations conducted into advanced production routes for the fabrication of porous copper structures using different powder types. Porous copper structures are beneficial for several applications, such as heat sinks, air filtration, and catalysts. The study started with the use of two different types of powder particles (spherical and dendritic) for the production of porous copper structures using hydraulic pressing. The processing conditions examined in this study include powder type, compaction pressures, and concentrations of a pore-forming agent (polyvinyl alcohol or PVA). After compaction, the samples underwent a two-stage sintering process at specific temperatures. The study examined the morphology, porosity, and mechanical properties of the sintered samples. The analysis revealed that samples with a higher weight percentage of PVA demonstrated better consolidation and overlapping of copper powder particles, resulting in improved morphology. The highest porosity was achieved when the dendritic copper powder was mixed with the highest weight percentage of PVA. The hardness of the samples varied significantly due to their high porosity. Where the samples were prepared using spherical powders at high pressure, the highest hardness was observed. The study concluded that porous copper structures with porosity ranging from 14% to 26% can be effectively produced by controlling the compaction pressure and PVA concentration. Furthermore, this master's thesis examined the application of nanotechnology to enhance the optical absorption and conductivity of copper during the laser sintering process. Copper powders were mixed with different concentrations of carbon nanotubes (CNTs) and the optical properties of mixed powders were evaluated using spectroscopy. The Box-Behnken Design of Experiments methodology was used to optimize the infrared laser processing conditions for sintering. Spectroscopic analysis was conducted to evaluate the reflection and thermal absorption of the IR wavelengths by the Cu-CNT composites. Density and hardness measurements were taken for the laser-sintered Cu-CNT pellets. The coating of copper powders with CNTs demonstrated enhanced optical absorption, resulting in reduced reflection. Due to the enhanced optical absorption, increased control and sensitivity of the laser sintering process were achieved, which enabled improvement in the mechanical properties of strength, hardness, and density, while also enabling control over the composite thermal expansion coefficient. A maximum average hardness of 66.5 HV was achieved. Indentation test results of the samples revealed maximum tangential and radial stresses of 0.148 MPa and 0.058 MPa, respectively. Overall, the thesis provides detailed insights into the production of porous copper structures and the potential benefits of incorporating CNTs for enhancing optical and material properties
Enabling Robust Automatic FAIRness Evaluation of Knowledge Graphs
Get PDFA knowledge graph is a form of knowledge representation that provides a mechanism for describing the interrelatedness of entities in a dataset.
Large knowledge graphs have become increasingly important in AI due to their ability to formalize and classify knowledge, enabling more effective extraction, retrieval, and analysis. They are used extensively in systems such as Google’s Gemini and Bard, and IBM’s Watson platform, to support smarter, context-aware applications in search, recommendation, and decision-making. As AI becomes part of daily life, the ethical implications of how these systems understand, recommend,
and decide are significant—making the reliability of their underlying data critical. A key consideration for any dataset is its adherence to the FAIR (Findable, Accessible, Interoperable, Reusable) principles, which aim to ensure the provenance, persistence, and reusability of data. In this context, FAIRness has become a crucial measure for establishing the suitability of knowledge graphs not only for data reliability but also for model reliability in machine learning. This thesis evaluates the three currently available automated tools for assessing knowledge graph FAIRness—F-UJI, FAIR Evaluator, and FAIR Checker—to determine their capabilities and consistency. These tools, while gaining adoption in
academic and industrial settings, have not previously been systematically compared. This work applies statistical analysis to assess the consistency of their outputs and finds that, while each tool has strengths, none alone offers a complete view. It proposes a novel consistency measurement to support complementary use of all three tools.
The systematic evaluation of FAIRness assessment tools, along with the introduction of a new supporting metric, contributes to more trustworthy knowledge graph assessments. This, in turn, provides a foundation for practitioners and researchers working in dataset curation and machine learning model development, where ethical and technical robustness are increasingly essential
Integrating people and technology: the impact of service system interactions on frontline work, customer outcomes, and organisational performance
No full textIn today’s dynamic service industry, organisations face the challenge of effectively integrating artificial intelligence (AI) and service-oriented human resource systems (SHPWS) to optimise frontline employee (FLE) service delivery and organisational performance. These elements are often implemented in silos, neglecting their systemic impact. This disconnect, coupled with the potential unforeseen consequences of FLEs’ perceptions of these integrated systems, poses a significant challenge.
Socio-technical systems theory advocates for a holistic approach to organisational design, recognising the interdependence of human and technological components. However, evidence linking this approach to improved organisational performance is limited. At the individual level, attributional theory offers a valuable lens for explaining how FLEs’ perceptions of workplace initiatives shape their attitudes and behaviours. Yet, it is often oversimplified, and narrowly applied to HR-related attribution triggers.
This study examines data from multiple sources in 50 hospitality settings across multiple countries at two time points. By modelling the interaction between intended AI actions and enacted SHPWS, it explores how FLEs perceive and react to these systems. At the individual level, FLE attributions (n=603) play a key role in shaping well-being and service performance. At the organisational level (n=50), findings highlight a complex relationship between enacted SHPWS and AI systems. Moderate SHPWS implementation generates positive synergies, while excessive reliance on technology can have adverse effects. Furthermore, the study identifies how differentiation strategy influences the relationship between FLEs’ attributions of service systems and their well-being.
This research highlights the potential of SHPWS – AI system integration to enhance FLE well-being and organisational performance. By examining the complex interactions between employees, customers, and organisations at the service frontlines, this study proposes a novel framework linking FLE attributions of service systems to performance outcomes. It offers actionable insights for organisations aiming to effectively balance human and technological elements in their service delivery
‘Chemnection’ An Interpretative Phenomenological Analysis of intimate relationships in the context of chemsex.
Get PDFThe term ‘chemsex’ describes the use of psychoactive drugs between gay and bisexual men who have sex with men (GBMSM) lasting several hours or days with multiple sexual partners. Chemsex is seen as enhancing sexual pleasure and fostering relationship growth, with shared experiences within social networks influencing participation and identity. The use of drugs facilitates emotional bonding and connection, offering affirming and sometimes transformative experiences. For many, chemsex fills a void of intimacy. This research builds on these insights, further exploring the relationship between chemsex and intimate relationships in ways that have yet to be fully examined.
Using Interpretive Phenomenological Analysis (IPA), the study investigates how participants define and navigate intimacy within the unique subcultural context of chemsex. Through in-depth qualitative analysis, eight participants’ subjective experiences were explored, uncovering patterns that illuminate the relational and interpersonal dimensions of chemsex. Three core themes appeared: (1) The Pressure of Intimate Relationships; where relationships, typically seen as sources of emotional support, were often described as overwhelming, enigmatic, and at times, suffocating; (2) Between Being and Nothingness; where chemsex is perceived by some as a space for temporary liberation, self-expression, and pleasure, while others experience it as destabilising, leading to loss of control and negative consequences, and (3) Seeking Acceptance; highlighting chemsex as a means of addressing deep-seated loneliness through a unique form of intimacy, albeit with ambivalent outcomes.
This research contributes to the growing literature on chemsex by offering a nuanced understanding of the complex emotional and relational dynamics within these encounters. The findings highlight the dual nature of chemsex as both a source of connection and disconnection, intimacy and isolation, self-realisation and depersonalisation. These insights provide valuable recommendations for psychotherapeutic interventions aimed at supporting GBMSM engaged in chemsex, enhancing therapeutic approaches to address the specific relational and emotional needs of this population
Visualizing stress granule dynamics with an RNA guanine quadruplex targeted ruthenium(II) peptide conjugate
Get PDFStress granules (SGs) are membraneless ribonucleoprotein assemblies that form in response to cellular stress. They have been linked to cell survival and cancer progression, though many questions remain
regarding their structure, function and therapeutic potential. Live-cell fluorescence imaging is key to advancing understanding of SGs, but there are very few small-molecule probes reported that selectively
image these organelles. RNA G-quadruplex (rG4) folding is believed to play a role in initiation of SG formation. Thus, to create a probe for SGs, we conjugated a G4 binding domain peptide from RNA helicase associated with AU-rich element (RHAU) to a luminescent [Ru(bpy)2(PIC-COOH)]2+, Ru-RHAU. Ru-RHAU is designed to target rG4s and thus SGs in live cells. Studies in cellulo demonstrate that Ru-RHAU can induce SG formation in a concentration and time dependent manner and immunolabelling confirmed the complex remains associated with rG4s in the SGs. The SG stimulation is attributed to stabilization of rG4 by Ru-RHAU consistent with rG4’s role in SG formation. Ru-RHAU
shows low cytotoxicity under imaging conditions, facilitating prolonged observation in live cells.Interestingly, under more intense photoirradiation, Ru-RHAU induces phototoxicity through an apoptotic
pathway. Ru-RHAU is a versatile tool for probing SG dynamics and function in cellular stress responses
and has heretofore unconsidered potential in phototherapeutic applications targeting SGs
Transforming Justice Responses to Non-Recent Institutional Abuses
Get PDFTransforming Justice Responses to non-recent institutional abuses
by Anne-Marie McAlinden, Marie Keenan and James Gallen offers a
thoughtful and detailed analysis of the historical and cultural contexts in
which non-recent institutional abuses occurred in Ireland—North and
South—and the barriers to justice experienced by victims/survivors locally, nationally and internationally. Ireland is presented as a case study for thorough empirical and theoretical consideration of abuses by the church and state. The authors reflect on the challenges presented by national and international legal, policy and institutional responses to non-recent abuses, drawing on a body of literature on transitional justice, restorative justice, human rights and public law, as well as the wider criminological and victimological scholarship
Towards Non-Invasive Health Diagnostics – Profiling the Human Skin Volatile Emissions
Get PDFHuman skin constantly emits volatile organic compounds (VOCs) reflecting underlying biochemical processes, resident microbial activity and environmental influences. Current literature highlights need for standardisation of skin VOC collection and analysis. The myriad
of approaches employed adds to variability of skin VOC profiles reported. Addressing this, the present work explores skin derived VOCs, where Chapter 1 highlights interplay amongst
metabolic processes, microbial activity, environmental conditions and their impacts on recovered VOC profiles.
Chapter 2 explores skin VOC profile stability within and across different participants over time, towards understanding factors influencing healthy skin VOC profiles. Headspace solid-phase microextraction (HS-SPME) fibres were used for non-invasive volar forearm VOC
collection with gas-chromatography mass-spectrometry (GC-MS) analysis. Biophysical parameters (skin surface pH, tissue dielectric constant, transepidermal water loss) were collected alongside VOC samples to investigate correlations. Resulting data highlighted
stability within a single participant over the sampling duration whilst showing a greater variance between different participants.
Chapter 3 introduces a new skin VOC sampling method using planar films (instead of SPME fibres) with solvent extraction for GC-MS analysis. Merits include ‘cost-effectiveness’ and ‘ease of use’ compared to HS-SPME unit used by our group to date. Three free-standing
sorbents (polydimethylsiloxane, Tenax-TA, divinylbenzene) were explored for their ability to retain VOCs from gas HS containing VOC standards. Skin VOC collection using these samplers was explored by integrating these into a wearable format and applying on volar
forearm.
Chapter 4 evaluates conclusions from this thesis, summarising key findings and offering fresh perspectives for this field. This work provides foundation for evaluating VOC capture material performance, with high specificity and reproducibility, whilst addressing
critical gaps in current skin VOC workflows. Furthermore, this work introduces a wearable and scalable skin VOC collection platform, allowing decentralised VOC collection, towards increasing skin VOC study cohorts for high quality results within this emerging field