151398 research outputs found
Sort by
SAPP: Student Academic Performance Predictor
No full textPost COVID-19 pandemic, the education system has significantly shifted towards blended and hybrid learning approaches. In most cases, course delivery and materials are now being hosted online through various Learning Management Systems (LMS). In the recent years, researchers in the education paradigm have proposed methodologies to find key performance indicators from the LMS data and used Machine Learning (ML) models to identify students who require an early intervention to improve their academic performance. Although these ML-based solutions show high accuracy, these solutions are often restricted to a single course/module belonging to a specific department or university. Moreover, they evaluate a fixed set of ML models to determine the best performer. To address this, we propose a dynamic Student Academic Performance Predictor (SAPP) tool which can work for different types of modules having diverse student records. The tool can predict poorly performing students to make early intervention and improve their academic performance. As proof of concept, the tool has been designed as a Python application which can take LMS data from different modules and predict a list of students requiring early intervention. The application uses the best performing ML model out of 18 ML models. Initial results using LMS data collected from two different modules running at a UK university give us early indication that the SAPP tool can accurately predict student performance for various modules with diverse range of students
Incorporating Span 80 surfactant into lipid nanocapsules improves their biocompatibility and cellular uptake in B16F10 melanoma cells
No full textSurfactant-shell lipid nanocapsules (LNCs) are promising skin delivery systems. They are composed of an oily core with a stabilising shell of surfactant and phosphatidylcholine. LNCs’ hydrodynamic diameter can be easily tuned by varying the surfactant content in the formulation. Hydrophilic surfactants incorporated into LNCs have shown toxicity in mammalian cells. To date, the toxicity of all published surfactant-shelled LNCs produced by the the phase inverson temperature (PIT) method has been investigated using hydrophilic surfactants, with no studies examining the impact of incorporating hydrophobic surfactants on LNCs’ in vitro behaviour. Span 80 is a hydrophobic surfactant and has been extensively used in manufacturing various ranges of nanoparticles. The present study formulated Span 80-containing LNCs to evaluate their in vitro behaviour in the B16F10 melanoma cell line. LNC-100-S8 of Kolliphor HS15/Span 80 (65/35 w/w%) and original LNC100-0 LNCs of Kolliphor HS15 with a hydrodynamic diameter of 100 nm were prepared using the PIT method. A salt aggregation test confirmed increased surface hydrophobicity of LNC100-S8 compared to LNC100-0. Cytotoxicity assays demonstrated that LNC100-S8 had a three-fold lower cytotoxicity than LNC100-0 (IC80 = 11757 μg/mL vs 3184 μg/mL). Flow cytometry analysis indicated significantly higher cellular uptake of LNC100-S8 compared to LNC100-0, with 1.52-fold, 1.46-fold, and 1.67-fold increase at 1 h, 3 h, and 24 h, respectively . Mechanistic investigations revealed that LNC100-S8 uptake predominantly occured via phosphoinositide 3-kinase (PI3K)-regulated macropinocytosis and actin-dependent endocytosis, whereas LNC100-0 also utilised Na+/H+ exchanger-mediated macropinocytosis. Furthermore, protein corona analysis demonstrated increased interactions between LNC100-S8 and B16F10-conditioned media proteins, leading to bimodal size distribution and elevated polydispersity index (>0.3), which influenced their endocytic pathways. Overall, Our findings revealed the high promise of our Span 80-containing LNCs as a drug delivery system with enhanced cellular uptake and biocompatibility in B16F10 melanoma cells compared to conventional LNCs composed of Kolliphor HS15 surfactant, highlighting their potential uses in topical delivery to melanoma and other skin diseases.<br/
Developing a conceptual model for collective patterns and leisure-time physical activity inequalities amongst adults
No full textBackground: Despite extensive efforts to tackle the unequal levels of leisure-time physical activity (LTPA) among adults, our understanding of how these inequalities are sustained over time is minimal. Existing approaches often overlook the complex dynamic, and typically nonlinear interactions, which shape physical activity patterns. Therefore, this paper aims to capture these dynamic interactions and processes which influence an individual’s decision to practice LTPA, with a specific focus on the influence of income. Methods: The conceptual model was developed in 4 main stages: (1) initial draft development, (2) realist reviews, (3) expert consultation, and (4) finalization. Results: The conceptual model has intention as its central construct to practice LTPA, and intention is directly influenced by dynamic interrelated processes between individual’s economic, psychosocial, and perceived physical environment factors. Conclusion: From the multiple theories, evidence, and expert agreement, we are confident that our conceptual model expresses the dynamic interactions which shape an individual’s decision to practice LTPA in a context of socioeconomic inequalities. Going forward, this work will be used to conceptualize and inform the development of an agent-based model to simulate LTPA inequalities among adults and can be expanded or adapted to other uses, such as intervention development
Enhanced efficiency assessment in manufacturing: leveraging machine learning for improved performance analysis
No full textThis paper introduces EATBoosting, a novel application of gradient tree boosting within the Data Envelopment Analysis (DEA) framework, designed to address undesirable outputs in printed circuit board (PCB) manufacturing. Recognizing the challenge of balancing desirable and undesirable outputs in efficiency assessments, our approach leverages machine learning to enhance the discriminatory power of traditional DEA models, facilitating more precise efficiency estimations. By integrating gradient tree boosting, EATBoosting optimizes the handling of complex data patterns and maximizes accuracy in predicting production functions, thus improving upon the deterministic nature of conventional DEA and Free Disposal Hull methods. The practicality of our approach is demonstrated through its application to a PCB assembly process, highlighting its capacity to discern subtle inefficiencies that traditional methods might overlook. This methodology not only enriches the analytical toolkit available for operational efficiency analysis but also sets a precedent for incorporating advanced machine learning techniques in performance evaluation across various industries. Looking forward, the continued integration of such innovative methods promises to revolutionize efficiency analysis, making it more adaptive to complex industrial challenges and more reflective of real-world production dynamics. This work not only broadens the scope of DEA applications but also invites further research into the integration of machine learning to refine performance measurement and management
Benign Land (soundtrack for interactive audiovisual installation/game)
No full textComposition of seventeen interactive musical soundworlds for digital artwork 'Benign Land'.'Benign Land' is an interactive digital artwork by Netherlands based artist Leandros Ntolas, initially presented as an immersive interactive installation the imposing site of Carlisle Memorial Church, and subsequently published for public release as a video game on Steam. Ntolas has created an evocative installation that incorporates his long standing research into perception, architecture and light, and responds to the specificities of the site, while expanding his current research into creating interactive experiences using the medium of video games. The installation culminates in an interactive gaming experience on a large screen that invites the viewer to inhabit a fictional landscape that interweaves layers of history, folklore, literature and music relating to the land. The soundtrack composed by John D’Arcy features a series of seventeen contrasting soundworlds that draw on varying forms of traditional Irish music and literature, each located in one of the seventeen virtual 'lands' within the game. Benign Land is part of the Belfast 2024 was commissioned by Belfast City Council, British Council, and the Mondriaan Fund. It is kindly sponsored by Visual Spectrum Studios and made possible through the support of The Belfast Buildings Trust
A systematic review of the experiences and perceived impacts of advocacy support for Looked After Children and Young People within child welfare services
No full textA protocol for a systematic review on the experiences and perceived impacts of advocacy support for Looked After Children and Young People within child welfare services. The protocol will explain in depth the background to independent advocacy within a child welfare setting and the significance in it and will outline what the author/s will do throughout each stage of the systematic review. This systematic review has been designed to provide a detailed analysis of all appropriate literature within the field of advocacy for looked after children within a child welfare setting. the protocol has laid out each stage, and timelines that the review will follow
What do we mean by advocacy? Definitions, values and practices in social work with children: a scoping review
No full textA protocol for a scoping review on 'what we mean by advocacy- definitions, values and practices in social work with children.' The protocol will explain in depth the background to advocacy within a child welfare setting and the significance in it and will outline what the author/s will do throughout each stage of the review. This scoping review has been designed to provide a detailed analysis of all appropriate literature within the field of advocacy children and young people within a child welfare setting. the protocol has laid out each stage, and timelines that the review will follow
A new prototype of Fibre Optic Sensor (FOSS) to monitor bridge scour
No full textBridges play a key role in the transportation network system and are vulnerable to various natural hazards. For bridges constructed over rivers, scour is one of the major causes of bridge failure. With the growing impacts of climate change, both the intensity and frequency of flood events are increasing, raising significant concerns regarding the safety and maintenance of bridge infrastructure. This study presents a method for remotely monitoring scouring which uses a Fibre Optic scour sensor based on Fibre Bragg Grating, named FOSS. The sensor comprises three sensing elements (fins), which are embedded at different depths. The system operates on the principle that, when erosion occurs, the movement of these fins generates a measurable response, indicating the depth of erosion. This paper presents the sensor's characteristics and shows preliminary findings from a field deployment on a bridge in the Northern Ireland region.<br/
Share Your Screen:A model for asynchronous healthcare e-Learning development
No full textPolicy Brief presented alongside a poster presentation at the SPHERE Conference 2025