6277 research outputs found
Sort by
Introduction
No full textIt was the worst of times: times of incredulity, despair, and darkness. Yet we emerged from that darkness into a season of light, the spring of hope, a new epoch of belief. While ours may not be ‘the best of times’ yet, the years during and following the COVID-19 pandemic sparked opportunities for new ways of thinking, doing, and being in higher education (HE)
The validity and reliability of the my jump lab artificial intelligence application
Get PDFJump height (JH) achieved in a countermovement jump (CMJ) has been suggested to allow for the monitoring of neuromuscular fatigue (NMF) and assessment of lower body power. Although force platforms (FP) are considered the gold standard for measuring CMJ height, they are expensive compared to mobile apps such as My Jump Lab (MJL). Therefore, this study aimed to assess the concurrent validity and agreement of the MJL app compared to a FP (ForceDecks [FD]) system and to determine its test-rest reliability. A convenience sample of 26 (n = 11 females and n = 15 males) recreationally active university sport students and staff (mean ± SD; age: 23.08 ± 6.33 years; mass: 72.85 ± 9.93 kg; stature: 176.63 ± 10.18 cm) participated in the study. Participants attended the laboratory for testing on two separate occasions, separated by one week. After a standardised warm-up, they completed three CMJs on each occasion, with CMJ height simultaneously assessed by the FD and MJL app. The MJL Artificial Intelligence mode showed a mean bias of 4.32 cm [95% CI: 3.4, 5.26] overestimation with 95% limits of agreement ranging from -3.33 cm [95% CI: -4.96, -0.85] to 11.98 cm [95% CI: 10.13, 13.41]. Both methods demonstrated minimal mean bias (FD = 0.61 cm [95% CI: -0.31, 1.37] and MJL = 0.25 cm [95% CI = -0.48, 0.98]) between sessions, and both showed a similar width to their limits of agreement, ranging ~7 cm about the mean bias. In summary, the MLJ overestimated CMJ height in this sample compared to the FD system, but both methods were reliable. Given the significant differences in cost for these two methods, teams on a budget may interested in trialling the MJL app.
Circulating microRNAs as potential diagnostic tools for asthma and for indicating severe asthma risk
Get PDFAsthma places a significant burden at individual and societal levels, but there remains no gold-standard objective test for asthma diagnosis or asthma severity risk prediction. MicroRNAs (miRNAs) are short non-coding RNA sequences that are attracting interest as biological signatures of health and disease status. We sought to construct serum miRNA panels that could serve as potential biomarkers to aid in the diagnosis of asthma and predict asthma severity. Thirty-five asthma-related miRNAs were screened in the serum of three patient groups (never-asthma, mild-asthma, and severe-asthma; n = 50/group) drawn from two well-characterised cohorts. miRCURY LNA technology was used, followed by GeneGlobe analysis. The associations of miRNA expression with clinical outcomes of interest and diagnostic value of the proposed miRNA panels were assessed. We identified an asthma diagnosis panel comprising upregulated miR-223-3p, miR-191-5p, and miR-197-3p (area under curve (AUC) = 0.813, sensitivity 76% and specificity 72%). Compared with mild-asthma individuals, we also identified an asthma severity risk panel comprising upregulated miR-223-3p plus downregulated miR-30a-5p, miR-660-5p, and miR-125b-5p (AUC = 0.759, sensitivity 78%, specificity 64%). Individual miRNAs showed associations with worse clinical asthma severity and impaired quality of life. miRNA panels with high sensitivity and specificity offer potential as biomarkers for asthma diagnosis and asthma severity
Learning to code in a conversion course with content co-creation
No full textTeaching programming to mixed-background learners poses the challenge of balancing the learning experience in the class. In this research, we investigated the impact of content co-creation teaching approach for a programming module in the master's conversion course. The content co-creation allowed students to create their own problem to solve using the programming skills they learn in class. The class test and questionnaire were used to evaluate the effectiveness of the teaching approach. The results showed that there were no significant differences in test score between students with a computing and non-computing background. The participating students had a strong positive opinion about the teaching approach
Two-stage machine learning for efficient network intrusion detection in software defined networks
No full textIntrusion detection in computer networks is vital to mitigate the growing number of internet-based attacks and machine learning (ML) is an ideal candidate for classifying malicious traffic. However, due to high traffic volumes, existing ML approaches can be too computationally intensive for packet level intrusion detection in practical networks; this important constraint is not considered by most ML based intrusion detection research. This paper proposes a novel two-stage machine learning based solution for intrusion detection leveraging software-defined networking (SDN). The proposed solution distributes the machine learning tasks between a centralized classifier in the SDN controller and classifiers deployed at edge SDN switches. The centralized classifier uses straightforward, low-data rate, flow statistics to identify traffic flows as benign/malicious/uncertain. This centralised classification is used to instruct edge switches, through OpenFlow, to either forward or drop the traffic for traffic that is highly probable benign/malicious and only pass uncertain traffic to a packet-based classier at the edge. The proposal is evaluated using real traffic flow measurements to show that the processing requirements of the hierarchical approach is two or three orders of magnitude less than existing, purely edge based, ML approaches
Beyond polarity: forecasting consumer sentiment with aspect- and topic-conditioned time series models
Get PDFExisting approaches to social media sentiment analysis typically focus on static classification, offering limited foresight into how public opinion evolves. This study addresses that gap by introducing the Multi-Feature Sentiment-Driven Forecasting (MFSF) framework, a novel pipeline that enhances sentiment trend prediction by integrating rich contextual information from text. Using state-of-the-art transformer models on the Sentiment140 dataset, our framework extracts three concurrent signals from each tweet: sentiment polarity, aspect-based scores (e.g., ‘price’ and ‘service’), and topic embeddings. These features are aggregated into a daily multivariate time series. We then employ a SARIMAX model to forecast future sentiment, using the extracted aspect and topic data as predictive exogenous variables. Our results, validated on the historical Sentiment140 Twitter dataset, demonstrate the framework’s superior performance. The proposed multivariate model achieved a 26.6% improvement in forecasting accuracy (RMSE) over a traditional univariate ARIMA baseline. The analysis confirmed that conversational aspects like ‘service’ and ‘quality’ are statistically significant predictors of future sentiment. By leveraging the contextual drivers of conversation, the MFSF framework provides a more accurate and interpretable tool for businesses and policymakers to proactively monitor and anticipate shifts in public opinion
Multiple knowledge depiction of digital twin-driven circular economy: concepts, integrated advanced technologies, triple bottom line of smart construction, and exploratory case studies
Get PDFOngoing global issues arising from rapid societal and economic development intensify resource extraction, ending in waste and emissions. Digital technology-empowered circular economy (CE) practices potentially encounter these issues, leading to sustainable development (SD). Digital twin (DT), a building block of emerging digital technologies, is extensively employed to automate and modernize construction phases and services. However, the broader research body reflects knowledge gaps and evidence inadequacies. This study seeks to disseminate awareness of DT and how its adoption reinforces confidence in CE practices, prospectively nurturing the way for triple bottom line (TBL) sustainability in construction. The study comprehensively and rigorously reviews DT deployment with a customized focus on building construction, delving into and analyzing the significance and bottlenecks. This research endeavors to apprise evidence of DT-driven CE uptake in the Kingdom of Saudi Arabia (KSA), which has launched Vision 2030 to steer its nation towards a vibrant society, ambitious nation, and thriving economy. A four-step methodology is adopted for exploratory research case studies with a literature review. Pertinent literature is collected, focusing on 2020–2024. Subsequently, multiple cases from the KSA are explored to corroborate the phenomenon under scrutiny. Key findings discovered that DT capabilities enhanced with other technologies and tools strengthen physical and cyber systems and data infrastructure. DT-navigated CE shares a plethora of social, economic, and environmental opportunities in residential, industrial, and commercial buildings. This study encompasses theoretical and practical implications. It offers comprehensive insights into visionary concepts for the research community and construction industrialists, nurturing their understanding and motivating them to implement DT for CE catalysis, resulting in a multitude of TBL sustainable advantages. Besides, it bridges the gap between literature and practical real-world practices. The developed interdisciplinary framework enhances DT application feasibility in construction services within the boundaries of architecture, engineering, construction, and procurement services
Sensitivity analysis of WRF-CMAQ model in predicting PM2.5 and O3 concentration in Peninsular Malaysia: 2019 transboundary burning smoke case study
Get PDFThe high PM2.5 concentrations significantly influence the air quality in the Maritime Continent region, especially in Peninsular Malaysia (PMY), which is affected by the annual burning season. However, the 2019 pollution case is unique due to the presence of a positive Indian Ocean dipole (pIOD) with a weak El Niño, which influenced the transport of pollutants toward PMY. This work aims to evaluate the ability of the numerical chemical weather prediction model (WRF-CMAQ) by performing a sensitivity analysis to reproduce the air quality during this event. Two model settings were studied: weather nudging and the burning emission amount of the fire inventory from NCAR (FINN). Three cases were established: 1) WRF-CMAQw (without nudging setting and with original fire emission), 2) WRF-CMAQn (with nudging setting and with original fire emission), and 3) WRF-CMAQa (with nudging setting and adjusted fire emission) to predict the PM2.5 concentration in PMY during the 2019 transboundary smoke event. The weather (temperature and wind profile) simulation results showed that WRF-CMAQa and WRF-CMAQn agreed up about 95 % and WRF-CMAQw agreed up to 93 % when compared with ground weather stations based on the statistical evaluation of correlation coefficient, bias, and error measures. For air quality, overall, WRF-CMAQa (87.23 %) demonstrated better performance compared to WRF-CMAQw (62.41 %) and WRF-CMAQn (78.72 %) in predicting the ground PM2.5. However, the diurnal prediction during the transboundary smoke event remains weak. For O3 concentration, the model performance agreement was quite low for all simulations. However, WRF-CMAQa could predict about 44.76 % compared to WRF-CMAQn (26.66 %) and WRF-CMAQw (41.90 %) in overall model performance, and all simulations managed to capture the diurnal trend of O3 when compared with ground observation station data. In conclusion, the sensitivity study on the weather and chemical prediction model, especially WRF-CMAQ, could help improve the air quality prediction system in PMY during the recurrence of transboundary smoke events
A showcase for the development of women's football in Africa? The 2023 FIFA Women's World Cup and the underrepresentation of women coaches
No full textThis chapter explores the uneven playing field for women coaches in African football. It provides an overview of the obstacles and barriers women coaches in Africa face, before establishing potential factors that could enhance opportunities for women in coaching – something which could, in turn, contribute to the overall growth of football across the continent. While gender discrimination may be a global issue in football, the chapter argues that it is vital to examine whether practical and cultural differences present additional challenges for aspiring women coaches in African nations. The chapter then proposes ways of addressing some of the core issues for developing women coaches in Africa, whilst highlighting considerations for administrators and policymakers
The German Peasants' War, visual culture and political subjectivation
No full textThis study examines the visual productions of the German Peasants’ War – pamphlets, banners, and clothing – to argue for the disruptive and radical visual legacy in which hierarchies and modes of subjection were overturned.Drawing on the author’s experience as a print maker and artist, the book offers a close and sympathetic analysis of the visual culture produced in this moment of war and revolt. Far from only being a matter of historical interest, these disruptive modes of visual production also resonate with contemporary debates about dissensus, populism, and political identity, especially in the work of Jacques Rancière. The refusal of these peasants (and mercenaries and some clergy) to remain in their place ruptured the visual field of power. It was also the repression of this popular eruption that was to shape conventional visual culture and politics as a reaction.<br/
Data sources:
Access Repository Dashboard
Do you manage Solent University Research Portal? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!