6277 research outputs found
Sort by
The association between physical outputs and match outcome across different playing styles for a professional second-tier football team across two complete seasons
Get PDFBACKGROUND: Elite-level football requires an array of physical, technical, psychological, and tactical skills. The aim of this study was to measure the association between physical outputs (distance, decelerations, accelerations) and the match outcome (win, draw, lose) in professional football. This research also examined whether the same association is influenced if a team adopts a possession or transition-based playing style.METHODS: Thirty-six elite-outfield football players from an English Championship team participated in the study during the 2020/2021 and 2021/2022 seasons using physical and event data collected from an English Championship club over the 2020-2021 and 2021-2022 seasons, this study conducted a univariate analysis of variance (ANOVA) and Hedge g effect size (ES) to measure the research aims.RESULTS: The results showed no significant differences were found between match outcomes for each physical output metric calculated. There was a trivial ES shown for all conditions except decelerations, with win/lose having a moderate ES (g=0.53). When playing a possession-based playing style there was no significant difference or non-trivial ES found for any physical output and match outcomes. When playing a transition-based playing style there was a moderate ES found for win/draw (P=0.38, g=0.90) and win/loss (P=0.98, g=0.64).CONCLUSIONS: This research provides important evidence for utilizing intense deceleration actions as a physical KPI during match play for teams adopting a transitional playing style. Subsequently, training interventions should be adopted to physically prepare players to complete and sustain intense deceleration actions during match play.</p
Towards transparent diabetes prediction: combining AutoML and Explainable AI for improved clinical insights
Get PDFDiabetes is a global health challenge that requires early detection for effective management. This study integrates Automated Machine Learning (AutoML) with Explainable Artificial Intelligence (XAI) to improve diabetes risk prediction and enhance model interpretability for healthcare professionals. Using the Pima Indian Diabetes dataset, we developed an ensemble model with 85.01% accuracy leveraging AutoGluon’s AutoML framework. To address the “black-box” nature of machine learning, we applied XAI techniques, including SHapley Additive exPlanations (SHAP), Local Interpretable Model-Agnostic Explanations (LIME), Integrated Gradients (IG), Attention Mechanism (AM), and Counterfactual Analysis (CA), providing both global and patient-specific insights into critical risk factors such as glucose and BMI. These methods enable transparent and actionable predictions, supporting clinical decision-making. An interactive Streamlit application was developed to allow clinicians to explore feature importance and test hypothetical scenarios. Cross-validation confirmed the model’s robust performance across diverse datasets. This study demonstrates the integration of AutoML with XAI as a pathway to achieving accurate, interpretable models that foster transparency and trust while supporting actionable clinical decisions
A comprehensive analysis of alcohol-attributed mortality in the United States
Get PDFAlcohol related mortality remains an important public health challenge in the United States (US), with patterns that vary substantially by demographics and geography. The objective of this study is to provide a thorough analysis using state-of-the-art data mining and machine learning techniques on alcohol attributable deaths in the U.S. from 2015–2019. Driven by the urgent need for specific interventions, we used regression models followed with ensemble techniques (XGBoost) to forecast mortality rates utilizing parameters such as age, geographic region of residence, manner of death and consumption patterns. Results in accordance with empirical CDC mortality data, XGBoost outperformed all other models in predicting age-based (R2 = 0.98) and cause-specific (R2 = 0.96). Geographic patterns show that California, Texas and Florida were particularly hot spots for alcohol versus suicide related deaths. The study also found demographic disparities, with older adults being more at risk. Major research contributions are the good predictability of news on general mortality and due to age, selected causes of death as well identification these regional patterns. These findings are invaluable in shaping targeted public health initiatives and policies to address alcohol-related harm. The adoption of sophisticated predictive modelling methodology in this study contributes significantly to the field by providing an empirically driven process for investigating and confronting alcohol-attributable death burden within the U.S
Predicting health care facility stay duration: a machine learning approach
Get PDFThe COVID 19 pandemic revealed shortcomings in healthcare, particularly concerning bed occupancy and resource allocation. During the Delta variant wave, it was highlighted how much improvement is needed in management strategies. One promising solution is the prediction of inpatient Length of Stay. Accurate predictions can enhance efficiency, reduce infection risks, lower mortality rates and decrease bed occupancy. This research proposes a predictive model using Random Forest Regression to accurately forecast hospital length of stay, aiming to enhance resource management and patient care. We utilized a 2010 inpatient dataset from the New York Department of Health and conducted thorough data preprocessing, including cleaning, handling missing values, and numerical encoding of categorical variables for regression. Additionally, we experimented with three database variations: one with targeted and frequency encoding, another using synthetic minority oversampling technique for handling imbalances, and a third applying synthetic minority oversampling technique for regression with gaussian noise for continuous variables. Each database was tested with and without scaling using four different scalers. The objective was to achieve a mean absolute error below the industry standard of 6.5, prioritizing unbiased metrics. Our results indicate that the final model achieved a 2.93 mean absolute error on the normal database, demonstrating its effectiveness in predicting length of stay. The study underlined the potential of machine learning in accurately predicting the Length of Stay in hospitals and the possibility of a more accurate model of the industry standard. Further advancements could be made to the models with more balanced datasets and a user-friendly interface for hospital staff usage
Faith veganism: how the ethics, values, and practices of UK-based Muslim, Jewish, and Christian vegans reshape veganism and religiosity
No full textAccurate multilevel thresholding image segmentation via oppositional Snake Optimization algorithm: Real cases with liver disease
No full textLiver-related diseases significantly contribute to global mortality rates. Accurate segmentation of liver disease from CT scans is essential for early diagnosis and treatment selection, particularly in computer-aided diagnosis (CAD) systems. To address challenges posed by inconsistent liver presence and unclear boundaries, an enhanced Snake Optimization (SO) algorithm is proposed that integrates with opposition-based learning (OBL) called (SO-OBL), proving effective in global optimization and multilevel image segmentation. Experiments using CEC’2022 test functions compare SO-OBL with eleven recent and state-of-the-art metaheuristic algorithms, demonstrating its superior performance. Additionally, an advanced liver disease segmentation model based on SO-OBL incorporates an optimized multilevel thresholding technique, leveraging Otsu’s function. Notable segmentation metric results, including FSIM = 0.947, SSIM = 0.941, PSNR = 24.876, MSE = 236.88, and execution time = 0.281, underscore the model’s efficiency and potential for accurate diagnosis in CAD systems
"Where do we belong?" Collaborative insights from RAISE Special Issue Groups' (Early Careers and Research Evaluation) Writing Project
No full textRAISE Special Interest Groups (Early Career Researchers and Research & Evaluation) collaboration: A case study
No full textIn the academic year 2022/23, the RAISE Special Interest Groups for Early Career Researchers and Research &amp; Evaluation collaboratively developed a professional development programme for HE colleagues new to writing about student engagement. The diverse audience ranged from Early Career Researchers (ECRs) to colleagues new to academic writing including those interested in writing about Student Engagement. The programme featured three online events (alongside virtual on-demand support) covering themes around barriers and challenges to publication; enabling collaboration and co-creation across institutional/disciplinary contexts and the opportunity to participate in an academic writing workshop. This case study will present an account of the process and experiences of delivering these events looking into the barriers/challenges experienced by ECRs, the community-based, peer-learning approach adopted (CoPs) to address these with the aim to facilitate the publication process and make it more inclusive and accessible for (a diverse range of) participants. The example is framed and contextualised through relevant literature and a wider higher education backdrop of work-life balance, principles of staff-student partnership and a 'publish or perish' culture
Tactile transformation in flying airplanes: from hands-on to fingers-on aviation
No full textPortable laptops, cell phones, touchscreen equipment and other mobile devices are changing the way commercial airplane pilots are handling information used for flying aircraft. Pilot expertise and skill are being transformed by a new approach in which fingertips are replacing traditional hands-on methods of controlling airplanes. Drawing on participatory and interview methods at a UK airbase, this article draws on ethnographic research with commercial pilots and pilot cadets, to trace the refashioning of cell phone media in an aviation context where touch-based computer screens replace traditional airplane technology. Drawing on Merleau-Ponty’s phenomenology, this article examines how the knowing, sensing and intuition through the hands allows for a particular sort of la prise (grip) which accustoms itself to the emergence of new tactile and human–computer interfaces
The opportunities, challenges, and rewards of ‘community peer research’: reflections on research practice
No full textThis paper shares reflections from a group of academic researchers at the same University on their experience of conducting ‘community peer research’ projects involving non-academics in social research. We review a range of literature that has influenced the development of our practice, stressing the importance of co-production and power relations. We present six case studies that represent the breadth of our different engagements with community peer research, and then go on to reflect on the challenges and benefits of this approach. We identify a number of practical challenges, ways in which we overcame them, and in particular stress the importance of providing well-designed training for community peer researchers. We conclude with some recommendations for other researchers looking to conduct similar research
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!