6277 research outputs found
Sort by
Enhancing maritime education for digital sustainability
No full textPurpose Digital sustainability involves the ability of industries and professionals to adapt to rapidly changing technological landscapes. Digitalisation and artificial intelligence (AI) are expected to radically change the maritime industry’s job landscape, especially with autonomous ships. International organisations currently do not formalise the education of maritime professionals and deck officers and need new formal modules. This study aims to contribute to this aspect by investigating learners’ experiences and knowledge gaps in the fundamentals of, as supported by the andragogy theory, topics such as computer programming, cybersecurity and statistics. Design/methodology/approach The research was carried out at Southampton Solent University, with samples of 105 students attending various MSc courses in maritime operations and deck cadet courses. The data was collected through an online survey. The two groups were compared and analysed using a chi-square test. Findings The results show that the percentage of MSc students with previous training in statistics, computer programming and cybersecurity courses was 37%, 13% and 16%, respectively. The deck officers’ training in the same areas was 06%, 09% and 09%. The results of this study were used to develop a new maritime digital module to focus on these topics. Originality/valueThe paper highlights digital sustainability’s significance in adapting education and training courses. Ship management companies and higher education institutions must urgently meet the demands of digitalisation and AI in the maritime industry. It highlights the necessity of addressing current knowledge gaps and implementing new educational modules to ensure the sustainable development of digital skills among maritime professionals and cadets
Exploring team dynamics through network analysis: a season review of an elite Portuguese soccer team
No full textSocial network analysis was applied to investigate team dynamics and inter-player connections during matches to offer deeper insights into the organizational framework of an elite soccer team competing in the Portuguese First Division during the 2020–2021 season. This study aimed to assess the impact of match outcomes and the deployment of various tactical systems on the team’s macro network metrics, such as density and clustering coefficients. Data was collected from thirty-four matches, with each match’s passing interactions meticulously analyzed to construct adjacency matrices, thereby quantifying player interconnections. The study’s findings revealed a nuanced relationship between network metrics and match outcomes. Density was significantly higher in matches that ended in losses, suggesting a potential over-reliance on certain players or interactions in adverse scenarios. Conversely, matches won were characterized by higher clustering coefficients, indicating a more cohesive and interconnected team effort. The analysis of five different tactical systems revealed significant differences in density, pointing to the influence of tactical choices on player interactions. No significant differences were found in clustering coefficients across the tactical systems, suggesting a consistent internal team cohesion irrespective of the strategy employed. These insights highlight the utility of network analysis in enhancing the understanding of team dynamics and strategic planning. This study underscores the potential of such analytical approaches to inform better tactical decisions and optimize team performance, ultimately contributing to a more sophisticated level of competitive analysis in professional sports
Mitigating fuel station drive-offs using AI: YOLOv8 OCR and MOT History API for detecting fake and altered plates
No full textFuel station drive-offs, wherein the drivers simply drive off without paying, are a major issue in the UK (United Kingdom) due to rising fuel costs and financial hardships. The phenomenon has increased greatly over the last few years, with reports indicating a substantial increase in such events in the major cities. Traditional prevention measures such as Avutec and Driveoffalert rely primarily on expensive infrastructure and blacklisted databases. Such systems typically involve costly camera installation and maintenance and are consequently out of the budget of small fuel stations. These conventional approaches also fall short regarding real-time recognition, particularly regarding first-time impostors using fictitious plates, which represent an increasingly significant proportion of such forgery. This research presents an AI (Artificial Intelligence)-driven detection system using the MOT (Ministry of Transport) History API(Application Programming Interface) to scan in real-time at gas stations to recognize and prevent such fraud. The system integrates various state-of-the-art technologies to offer a foolproof system. Using the latest YOLO (You Only Look Once)model to recognize number plates and Easy OCR (Optical Character Recognition) to recognize characters, the system correctly reads license plates in various environmental conditions like lighting, viewpoint, and weather conditions. This approach minimizes the utilization of expensive camera systems and employs cheaper ANPR (Automatic Number Plate Recognition) gear, availing existing installed surveillance cameras on filling stations. The system operates with a basic web-based application to notify operators of stolen vehicles in real-time, enabling them to react immediately. Real-world testing achieves 84% success with CCTV (Closed-Circuit Television) images, depicting its real-world applicability. The results indicate that the AI-driven solution offers a monumental leap compared to current practices, giving fuel stations a cost-effective and efficient means of reducing financial loss from drive-off incidents
Machine learning approaches to cryptocurrency trading optimization: a comparative analysis of predictive models
No full textCryptocurrency markets are characterized by high volatility and complex patterns, creating both challenges and opportunities for traders and investors. This study introduces a machine learning framework for cryptocurrency trading optimization that leverages advanced analytical techniques to enhance trading decisions. We extracted historical data for 30 cryptocurrencies over a four-year period from Yahoo Finance. After preprocessing, we applied Principal Component Analysis (PCA) and K-means clustering to select representative coins. Four machine learning models (Gradient Boosting, XGBoost, Support Vector Regression, and Long Short-Term Memory networks) were trained to predict cryptocurrency price movements. Model performance was evaluated using multiple metrics, including Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and R-squared (R2). Gradient Boosting and XGBoost consistently outperformed SVR and LSTM models across all cryptocurrencies, with R2 values of approximately 0.98 for most coins. The framework successfully identified trading signals through both moving average strategies and machine learning predictions, providing actionable insights for cryptocurrency traders. Our analysis demonstrates that ensemble-based models offer superior performance for cryptocurrency price prediction compared to neural network approaches. The integration of advanced visualization tools and trading signal generation creates a comprehensive system for data-driven cryptocurrency trading decisions
Knowledge production as identity expression: a third-space perspective on sustainable careers
No full textThis conceptual paper advances our understanding of sustainable careers by focusing on the knowledge creation activities of third-space practitioners in higher education. Specifically, we apply and extend the Process Model of Sustainable Careers to examine how different forms of knowledge production in three different institutional and national contexts ? writing for publication, practice-based expertise, and curriculum development ? contribute to career sustainability at the intersection of academic and professional/administrative roles and domains. The critical element of this extension is ?identity expression? in the form of knowledge production, which acts as an integrating mechanism that flows through and connects the original model?s three dimensions of person, context and time. The paper responds directly to the calls for research on conceptualising sustainable careers by showing how varied knowledge production activities enable third-space practitioners to navigate the complexities of career pathways in boundary-spanning roles
Key success factors for adoption of CI/CD with agile project management - systematic literature review
No full textThe adoption of Continuous Integration and Continuous Delivery (CI/CD) is increasingly critical for improving software development efficiency and quality, especially in Agile Project Management. However, limited research focuses on the success factors driving CI/CD adoption in Agile environments. This systematic literature review identifies key factors for successful CI/CD integration, such as Customer engagement, Effective communication, Measurement, Collaborative Organizational Culture, Team Roles and Dynamics, Top management support, Tools, Continuous Monitoring and Continuous Improvement, Critical Skills and Employee Engagement. The review also highlights challenges such as resistance to change, lack of expertise in CI/CD tools, and difficulties in fostering cross-functional collaboration. Addressing both technical and organizational challenges is essential for successful CI/CD adoption. The review emphasizes the need for further research, particularly in customized and project-based environments, to better understand CI/CD adoption complexities and provide actionable insights for organizations
Beyond polarity: forecasting consumer sentiment with aspect- and topic-conditioned time series models
No full textExisting 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
Social network analysis in football: a systematic review of performance and tactical applications
No full textIntroduction: This systematic review aims to critically examine the application of social network analysis (SNA) in football, with a focus on its contribution to evaluating team performance, tactical behavior, and player interactions. Methods: Following PRISMA guidelines, a comprehensive search was conducted across four databases (PubMed, Scopus, Web of Science, and SPORTDiscus) from January 2017 to October 2024. Results: Fifty-five peer-reviewed studies met the inclusion criteria, addressing network analysis in official men's professional football matches. Data were extracted and summarized regarding methodological quality, network metrics used, tactical context, and practical implications. Discussion: Most studies demonstrated that cohesive network structures, characterized by high density, clustering coefficients, and centrality, are associated with successful team performance. Centrality metrics were frequently used to identify key tactical players, typically central defenders and midfielders. Recent methodological advances included dynamic time-window analysis, pitch-passing networks, and spatial-temporal integration using tracking data. However, there remains an overrepresentation of elite men's football and offensive phases, with limited focus on defensive networks, youth categories, and women's football. SNA offers a powerful framework to decode the complexity of football performance, evolving from static graphs to dynamic, rolesensitive, and context-rich models. Future research should adopt longitudinal designs, multi-layer network approaches, and closer collaboration with practitioners to enhance the operational utility of network insights in coaching and performance analysis. Systematic review registration: https://osf.io/2pe3
Antecedents of the actual usage of HRIS by employees in WFH and hybrid contexts: integration of DIT, TAM, and UTAUT
Get PDFThis study integrates the diffusion of innovation theory (DIT), technology acceptance model (TAM), and unified theory of acceptance and use of technology (UTAUT) to explore the antecedents of the actual usage of human resource information systems (HRISs) by employees in work-from-home (WFH) and hybrid contexts. The study gathered a total of 274 usable responses from employees across ten Sri Lankan software companies. Data analysis utilized the partial least square (PLS) path modelling technique, chosen due to the non-normality of the data. The research findings revealed that observability influences the perceived ease of use (PEOU) of HRIS in WFH and hybrid contexts. Similarly, compatibility emerged as the most critical attribute affecting perceived usefulness (PU). Both PEOU and PU, in conjunction with social influence, contribute to driving employees’ behavioural intention to use HRIS in WFH and hybrid contexts. Ultimately, behavioural intention and facilitating conditions were identified as the major variables that impact the actual usage of HRIS by employees in WFH and hybrid contexts. Notably, within the expanding body of post-pandemic business literature, this study represents one of the very first to investigate the actual usage of HRIS in WFH and hybrid contexts. Practitioners can utilize the findings to enhance and optimize their HRIS adoption efforts
Mitigating fuel station drive-offs using AI: YOLOv8 OCR and MOT History API for detecting fake and altered plates
Get PDFFuel station drive-offs, wherein the drivers simply drive off without paying, are a major issue in the UK (United Kingdom) due to rising fuel costs and financial hardships. The phenomenon has increased greatly over the last few years, with reports indicating a substantial increase in such events in the major cities. Traditional prevention measures such as Avutec and Driveoffalert rely primarily on expensive infrastructure and blacklisted databases. Such systems typically involve costly camera installation and maintenance and are consequently out of the budget of small fuel stations. These conventional approaches also fall short regarding real-time recognition, particularly regarding first-time impostors using fictitious plates, which represent an increasingly significant proportion of such forgery. This research presents an AI (Artificial Intelligence)-driven detection system using the MOT (Ministry of Transport) History API(Application Programming Interface) to scan in real-time at gas stations to recognize and prevent such fraud. The system integrates various state-of-the-art technologies to offer a foolproof system. Using the latest YOLO (You Only Look Once)model to recognize number plates and Easy OCR (Optical Character Recognition) to recognize characters, the system correctly reads license plates in various environmental conditions like lighting, viewpoint, and weather conditions. This approach minimizes the utilization of expensive camera systems and employs cheaper ANPR (Automatic Number Plate Recognition) gear, availing existing installed surveillance cameras on filling stations. The system operates with a basic web-based application to notify operators of stolen vehicles in real-time, enabling them to react immediately. Real-world testing achieves 84% success with CCTV (Closed-Circuit Television) images, depicting its real-world applicability. The results indicate that the AI-driven solution offers a monumental leap compared to current practices, giving fuel stations a cost-effective and efficient means of reducing financial loss from drive-off incidents
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!