21649 research outputs found
Sort by
Being Life-wide: Case Studies of Empathy-based Pedagogy, Enhancing Learning Journey's in the Workplace
No full textA Lightweight Hybrid Deep Learning-Based Intrusion Detection System for Detecting Botnet Attacks in IoT Networks
Get PDFThe rapid expansion of the Internet of Things (IoT) has introduced an unprecedented number of interconnected devices, creating new opportunities for automation and data exchange but simultaneously increasing the attack surface for cyber threats. Among these, botnet attacks have emerged as one of the most severe threats, capable of compromising massive networks of IoT devices for malicious purposes such as Distributed Denial of Service (DDoS) and data exfiltration. This research proposes a Lightweight Deep Learning-based Intrusion Detection System (DL-IDS) designed to efficiently detect IoT botnet activities with high accuracy and minimal computational overhead. The proposed model integrates a hybrid Autoencoder–Long Short-Term Memory (LSTM) architecture that captures both spatial and temporal traffic features. The system preprocesses raw network data through feature selection, Min–Max normalization, and class balancing using SMOTE, before training on benchmark datasets such as BoT-IoT and UNSW-NB15. Evaluation results demonstrate that the model achieves an overall accuracy of 99.4%, precision of 99.2%, and AUC score of 0.997, outperforming traditional machine learning and hybrid deep learning baselines (CNN–LSTM, GRU-hybrid, and SVM-Ensemble). Furthermore, the lightweight design of the model ensures its deployability in real-time edge computing environments. Experimental validation on a Raspberry Pi 4 (8 GB) confirms that the proposed DL-IDS maintains low latency and minimal memory consumption, making it well-suited for IoT gateway and edge-level applications. This study concludes that lightweight deep learning frameworks can effectively balance detection accuracy and computational efficiency, contributing significantly to securing modern IoT infrastructures against evolving botnet threats
GResMark: A swin transformer-based watermarking framework with geometric attack resilience
Get PDFDeep learning-based digital watermarking technology plays a crucial role in copyright protection and has been extensively researched. Current CNN-based robust watermarking methods embed the watermark into the deep features of the image through convolution operations. Despite the success of CNN-based watermarking frameworks, their reliance on capturing features at fixed spatial positions makes it challenging to address desynchronization caused by geometric distortions. To solve the above problem, we propose a robust watermarking framework based on the Swin Transformer named GResMark. Intuitively, GResMark leverages the spatial self-attention mechanism of the Swin Transformer to capture global and long-range dependencies among image tokens. Since geometric attacks preserve these relationships, GResMark achieves improved adaptability and robustness against such distortions. Specifically, we design the Locally-enhanced Channel Attention (LeCA) Swin Transformer block and Frequency Channel Attention (FCA) Transformer block as the backbone of our framework, enabling more effective dependency modeling for robust watermarking. In addition, to mitigate the impact of invalid pixel values in the watermarked images generated during training, which exceed the valid range defined by the underlying data format, we introduce a constrained loss that enhances the quality of watermarked images. Experimental results demonstrate that GResMark outperforms existing state-of-the-art (SOTA) watermarking methods in terms of robustness to geometric distortions and watermark capacity. For geometric attacks, GResMark achieves an extraction accuracy exceeding 98 % while maintaining superior visual quality
Dynamics and sediment transport in a path-induced blowout in opposing wind conditions
Get PDFBlowouts play a critical role in reactivating coastal dunes by serving as sand transport corridors. While extensive studies have explored aeolian processes within blowouts under onshore and oblique wind regimes, less attention has been given to environments dominated by opposing wind directions. The Canet-en-Roussillon coastal dune (SE France) is subjected to both offshore and onshore winds, with a main blowout (B2) exhibiting a complex morphology altered by human foot traffic. This blowout was instrumented (anemometers, sediment sand traps, topographic surveys) during both offshore and onshore wind events. The study demonstrates that onshore winds, though less frequent, are the dominant morphogenic force, driving rapid landward elongation of the blowout. Multi-year analyses reveal that this elongation has facilitated the connection with a closed footpath, resulting in its elbow-shaped morphology. At event timescales, offshore winds induce minimal sediment transport due to vegetated fetch surfaces, while onshore winds promote significant sand transport and topographic variability via a bare sand fetch. Sand availability emerges as a critical factor modulating blowout evolution. The dual wind regime interacts with anthropogenic disturbances, sustaining the current morphology. Offshore winds transport sediment seaward, while the bifurcation of onshore winds by the footpath creates an internal accumulation zone, inhibiting depositional lobe formation and further elongation. These findings challenge the traditional understanding of blowouts as unidirectional sand transport corridors in dual wind environments and highlight the role of anthropogenic influences on their morphology. Further research is needed to determine whether these mechanisms are consistent across larger blowouts with higher sand volumes
An update on the Young People’s Landscape Study: learning from stakeholder voices about research priorities in young people’s PICU and inpatient mental health care
Get PDFIn a previous editorial in this journal, Foster (2024) made the case for why research to examine the current provision of inpatient mental health services for CYP is so important. Building on this call to action, the current editorial provides an update on the Landscape Study (https://tinyurl.com/2hpn43m9), a three-year national research initiative, which aims to map how care provision for CYP experiencing acute and complex mental health needs is organised, implemented and used across England and Wales. In particular, this editorial draws on the insights generated through stakeholder engagement to suggest implications for policy, service specifications, clinical practice, and future research
The European Security Research & Innovation Ecosystem: A systemic approach to EU security R&I
No full textThe European Security Research & Innovation Ecosystem (ESRIE), developed by the Engage2innovate (E2i) project, presents a bold new way of understanding and improving how EU-funded security research is conceived, delivered and translated into real-world impact.Rooted in a proven human-centred design approach and validated through practical application, the ESRIE model maps the full lifecycle of security R&I — from upstream policymaking to downstream implementation. It reveals the hidden linkages between project-level R&I processes and EU-level funding cycles, highlighting how strategic alignment and better engagement with end-users can dramatically improve uptake and effectiveness.More than a model, ESRIE is a call for a self-reinforcing ecosystem: one in which research outputs, practical insights, and operational learning continually inform future funding decisions, ensuring that security R&I evolves in response to real needs — not administrative logic. It offers policymakers and practitioners a powerful framework for bridging the persistent gap between innovation and implementation.This paper outlines the rationale, structure and application of the ESRIE model. It provides not only conceptual clarity but practical direction for those seeking to deliver useful and usable security innovations that are actually implemented and have impact across Europe
Which Metrics can I Monitor? Test-Retest Reliability of Countermovement Jump and Countermovement Rebound Jump Force-Time Metrics in Youth Soccer Players In-Season
Get PDFObjective measures provide the most effective means of monitoring the magnitude and time course of changes in neuromuscular function (NMF) resulting from physical activity if measurements are repeatable (reliability). The aim of this study was to determine the test-retest reliability of a range of ratio, outcome, strategy, and kinetic force plate metrics for the countermovement jump (CMJ) and countermovement rebound jump (CMRJ) tests in youth soccer players in the in-season period. A test-retest repeated-measures design was employed consisting of two testing sessions separated by 7-days. In each testing session, male youth soccer players (N = 43; age 17.9 ± 0.9 years, height 181 ± 5.8 cm, body mass 72.5 ± 6.8 kg) from full-time English Football League academies (categories 2 and 4) performed three maximal-effort CMJs and CMRJs in a randomised order on a Hawkin Dynamics Inc. force plate system sampled at 1000 Hz. Fifteen out of 25 CMJ, 11 out of 19 CMJ portion, and five out of 19 rebound jump (RJ) portion metrics demonstrated acceptable absolute (coefficient of variation [CV] ≤10%) and relative (intraclass correlation coefficient [ICC] ≥0.75) reliability for youth soccer players in the in-season period. The CMJ test is a feasible and reliable slow stretch-shortening cycle (SSC) test which can be utilised for monitoring acute changes in NMF in youth soccer players in-season. Practitioners should consider applying a combination of CMJ outcome, strategy, and kinetic metrics in their monitoring processes. The CMRJ test is a less feasible yet reliable fast SSC test which can be utilised for monitoring acute changes in NMF in youth soccer players in-season and can be considered an appropriate alternative to the drop jump (DJ) test as it overcame issues in D
Immersive Learning in Construction and Civil Engineering Education
Get PDFImmersive learning has gained increasing attention in construction and civil engineering education as institutions confront the demands of Industry 4.0 and the learning preferences of Generation Z. Although virtual, augmented, mixed, and extended reality (VR/AR/MR/XR) technologies have demonstrated strong potential to enhance experiential, interactive, and scenario-based instruction, their adoption remains fragmented and underdeveloped - particularly in developing countries such as Vietnam. This study conducts a PRISMA-guided systematic review of 33 peer-reviewed publications to examine global applications of immersive learning in construction and civil engineering higher education. The findings reveal six dominant thematic domains: safety and hazard recognition, design and visualization, construction methods and procedures, project management and collaboration, robotics and equipment training, and energy and environmental simulation. Building on these themes, the study proposes the Immersive Learning Integration Framework for Construction and Civil Engineering Education (ILIF-CEE), which organizes immersive learning into five dimensions: learning objectives, technology types, interaction and immersion mechanisms, learning outcomes, and implementation enablers and barriers. The framework provides a structured foundation for integrating immersive tools into Architecture, Engineering, and Construction (AEC) curricula and for supporting institutional digital-transformation strategies. The study contributes consolidated evidence based on the reviewed publications, identifies critical challenges, and outlines practical implications for educators and policymakers seeking to adopt immersive learning
Understanding Reproductive Coercion in Cults and Destructive Group Settings
Get PDFThis quantitative study is the first exploratory analysis of the complex experiences of individuals who have experienced reproductive coercion while under the influence of a cultic or destructive group. Unlike previous anecdotal accounts from former members, this research systematically investigates how reproductive coercion manifests within destructive group settings, thereby addressing a long-standing gap in the literature. Ninety-nine participants were recruited through a targeted online sampling of self-identified former cult members. Participants were at least 18 years old, identified as female during their time with the group, and experienced a lack of control over their reproductive choices during that period. They completed an online questionnaire that explored their identity, reproductive autonomy, and psychological abuse experienced while in a destructive group setting. In contrast to prior research on relationship control, which found male partners to be the primary source of reproductive interference, this study demonstrated that the group’s ideology exerted the most influence on reproductive decision-making. Results also suggest that individuals born and raised in cultic groups experience reproductive coercion differently than first-generation members. This study provides insight into how cults can disturbingly control even the most personal issues in individuals’ lives and can serve as a powerful means of coercive control
Comparative analysis of academic performance in business management education: foundation year vs. non-foundation students in UAE higher education
Get PDFThis paper reviews the performance of students who graduated from the foundation year programme in business studies and compares their performance with non-foundation students at the higher national diploma equivalent to the foundation degree in the business programme. The study is conducted to understand how foundation students perform at the same level as their peers who have completed their schooling from the traditional school system in UAE. Data collected was secondary from the institution’s archive files of both the groups of students and their academic performance over fifteen modules. Descriptive statistical techniques were used to analyse the data. The results from the analysis found that the foundation students performed equally better as non-foundation students in most modules while in a few cores mandatory modules, students performed better than non-foundation students. The foundation programme students were not found to underperform in any module. In specialisation modules the data varied, however these modules were in marketing/finance/business in which the results were no longer comparable due to differences in specialisation, the field of study, and the type of assessments varied. Foundation students exhibited greater variability in their grades. This study contributes to scarce literature and research work in the area of foundation education in the UAE higher education teaching and learning and guides the future direction of studies based on the findings
Data sources:
Access Repository Dashboard
Do you manage University of Salford Institutional Repository? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!