17628 research outputs found
Sort by
A hybrid vision transformer with ensemble CNN framework for cervical cancer diagnosis
Get PDFCervical cancer is the leading cause of cancer-related deaths among women worldwide, necessitating early and accurate detection methods. This study introduces a hybrid framework utilizing Vision Transformers (ViT) and ensemble learning-based convolutional neural networks (CNN) models for cervical cancer classification based on Pap smear images. Two prominent datasets, Mendeley LBC and SIPaKMeD, are employed, encompassing nine distinct categories of cervical cell abnormalities. The proposed approach integrates pre-trained CNN models of DenseNet201, Xception, and InceptionResNetV2 to extract high-level features, further fused through ensemble learning. These features are then processed by the ViT-based encoder model designed for improved interpretability and accuracy. Experimental results demonstrate that the hybrid model achieves superior accuracy rates of 97.26%, a recall of 97.27%, a precision of 97.27%, and 96.70% for the F1-score on the Mendeley LBC dataset. For the SIPaKMeD dataset, there was an accuracy of 99.18%, a recall of 99.18%, a precision of 99.15%, and a 99.21% F1-score. On the combined dataset, the model outperformed individual pre-trained models with 95.10% accuracy and a 95.01% F1-score. Moreover, the framework incorporates augmentation with Explainable AI (XAI) techniques, specifically Grad-CAM, to provide transparent and interpretable diagnostic outcomes, enhancing its utility in clinical settings. This research underscores the potential of hybrid AI frameworks in revolutionizing cervical cancer diagnostics by offering accurate, efficient, and interpretable solutions
Electrochemistry in Scottish secondary schools: exploring teacher confidence and student misconceptions
No full textTeachers across Scotland rank electrochemistry as their least confident topic in both National 5 (age 15–16) and Higher (age 16–17) courses. This research explores the challenges and misconceptions surrounding students’ comprehension of electrochemistry from the perspective of classroom teachers. Through critical analysis of electrochemical topics in the Scottish Curriculum for Excellence, in combination with examination statistics and survey data, rational conclusions can be derived on the roots of these challenges and how they can be addressed
A Fuzzy Jump-Diffusion Option Pricing Model Based on the Merton Formula
No full textThis paper proposes a fuzzy jump-diffusion (FJD) option pricing model based on Merton (J Financ Econ 3:125–144, 1976) normal jump-diffusion price dynamics. The logarithm of the stock price is assumed to be a Gaussian fuzzy number and the risk-free interest rate, diffusion, and jump parameters of the Merton model are assumed to be triangular fuzzy numbers to model the impreciseness which occurs due to the variation in financial markets. Using these assumptions, a fuzzy formula for the European call option price has been proposed. Given any value of the option price, its belief degree is obtained by using the bisection search algorithm. Our FJD model is an extension of Xu et al. (Insur Math Econ 44:337–344, 2009) fuzzy normal jump-diffusion model and has been tested on NIFTY 50 and Nikkei 225 indices options. The fuzzy call option prices are defuzzified and it has been found that our FJD model outperforms Wu et al. (Comput Oper Res 31:069–1081, 2004) fuzzy Black-Scholes model for in-the-money (ITM) and near-the-money (NTM) options as well as outperforms Xu et al. (Insur Math Econ 44:337– 344, 2009) model for both ITM and out-of-the-money (OTM) options
Geographic variations in place of death and palliative care utilisation in the last three months of life in high-income countries: a systematic review
Get PDFBackground: Geographic variations in end-of-life healthcare utilisation may highlight disparities in access to care, including palliative care. Studies demonstrate that place of death and palliative care utilisation are influenced by geographic exposures such as sociodemographic and socioeconomic characteristics, rurality, and health service availability. No systematic review has synthesised the evidence across multiple geographic exposures. This is important to provide conclusions from a broader evidence base and inform equitable commissioning of palliative care services. Aim To investigate the influence of geographic exposures on individual-level place of death and palliative care utilisation in the last three months of life, to identify potential disparities in end-of-life healthcare provision. Methods: In this systematic literature review of quantitative data with narrative synthesis, nine databases were searched for peer-reviewed observational studies published between 1st January 2004 and 1st October 2024. Eligible studies recruited adults in high-income countries and compared two or more geographic exposures. Methodological quality was assessed using the Newcastle-Ottawa Scale. Results: Of 9,296 studies identified, 51 retrospective studies across 24 countries were included. Rurality was positively associated with home death and decreased palliative care utilisation. Regarding healthcare availability, greater bed availability in hospital or long-term care facilities increased likelihood of death in that facility. Similarly, closer proximity to a hospice or hospital location increased likelihood of death in that facility. Hospital death may be positively associated with residence in certain countries, regions of high population density, and those containing capital cities. Conclusions: Findings highlight geographic variations as areas of focus for commissioners and policymakers to reduce local end-of-life healthcare inequities. We make recommendations regarding structural care gaps, service efficiency, and innovation in palliative care provision
Gamble in the Coral Sea: Japan's Offensive, The Carrier Battle, and the Road to Midway
No full textThe opening salvos of the Battle of the Coral Sea, the first large-scale carrier clash in history, were fired one month before Midway. Gamble in the Coral Sea recounts, for the first time in English, the story of this battle from the Japanese point of view. Based on extensive Japanese-language sources, author Michał A. Piegzik forcefully challenges established Western narratives surrounding this critical engagement in the Pacific War.Operation MO, the Japanese plan to seize Port Moresby, kicked off in early May 1942. By committing three carriers, including the famous Shōkaku and Zuikaku, the Nippon Kaigun’s command risked a critical part of their fleet just before the envisaged decisive battle at Midway in the Central Pacific, scheduled for early July. The operation was considered a vital part of Japanese strategy. Victory would isolate Australia and New Zealand and extend access to vital resources crucial to Japan’s war effort. Victory, however, would prove elusive after American codebreakers deciphered Japanese radio traffic that revealed their plans in the weeks leading up to the launch of Operation MO. Using this intelligence to their advantage, U.S forces located elements of the Japanese navy as they steamed through the Coral Sea. Soon after, history’s first carrier battle began. Piegzik combines expertise in military history with mastery of the Japanese language to provide a rare perspective on the Imperial Japanese Navy’s operational choices during the battle. His use of Japanese archival documents and personal testimonies from surviving Japanese crew members uncovers new dimensions to the battle. The clash proved to be a Pyrrhic victory for the Japanese, who sunk the Lexington and crippled the Yorktown but were forced to call off Operation MO due to the severe damage inflicted on Shōkaku and the heavy losses among their aircrews. Revealed here are the circumstances and actual reasons for the Japanese failure and the revised impact of the Battle of the Coral Sea on the Battle of Midway. Beyond tactical details, Piegzik offers insight into the broader consequences of the battle. He engages with sources previously underexplored and integrates them with Allied perspectives to ensure a well-rounded understanding of the events. A vital addition to any World War II collection, Gamble in the Coral Sea offers a nuanced and thorough exploration of a battle that significantly shaped the trajectory of the war in the Pacific
Impact of Chinese foreign direct investment on energy demand and stability: The Belt-Road Initiative as conditioner
Get PDFWe comparatively examine the impact of China's foreign direct investment (FDI) on energy demand for 29 Belt-Road Initiative (BRI) versus 25 non-BRI countries for 2000–2023. Employing partially-out LASSO linear regression and double selection LASSO linear regression, we find that FDI has a scale effect on BRI countries, while for non-BRI countries, it engenders more sustainability in production practices. Energy demand considerably escalates in the pre- and post-phase for BRI countries, with digitalization exerting no significant influence in either period. Conversely, in non-BRI countries, digitalization dampens the impact of FDI on energy demand in both the pre- and post-phase, reinforcing the sustainability effects of foreign investment
Knowledge Distillation for Lightweight and Explainable Intrusion Detection in Resource-Constrained Consumer Devices
Get PDFThe growing demand for deploying deep learning (DL) models for cyber-attack detection on consumer devices is constrained by their limited memory and computational resources. High-performing DL models are often too resource-intensive for on-device deployment. Knowledge distillation (KD) offers an alternative by transferring knowledge from a complex teacher model to a smaller, lightweight student model, reducing model size and inference time with minimal accuracy loss. In this paper, a KD-based framework is utilized to design and train both teacher and student models for intrusion detection using the N-BaIoT and CIC-IDS 2023 datasets. The student model achieves comparable performance to the teacher, with accuracies of 99.87% and 98.71% on the utilized datasets. It maintains accurate classification across various attack classes, achieving F1-scores above 0.98, while reducing computational complexity from 263k to 120k trainable parameters, model size from 1.00 MB to 471.67 KB, and inference time from 48 ms to 34 ms. Moreover, SHAP-based explainable AI techniques are employed to interpret both models, demonstrating that the distilled student model retains key feature attributions of the teacher model
Ethical Funds and Risk: Evidence from the United Kingdom
Get PDFThis thesis mainly aims to analyse the risk exposure of ethical funds. It also examines the risk-adjusted return (RAR) of ethical funds. In this study, Islamic funds and ethical funds are treated as distinct types of funds due to their underlying unique asset screening criteria. The research objectives are to compare the risk exposures of ethical funds and conventional funds; to examine the risk exposures across different ethical fund styles; to explore the risk exposures of Islamic funds in comparison to ethical and conventional funds, and to examine the RAR across different fund types. Thirteen distinctive risk measures and six unique RAR measures are used to quantify the riskiness and performance of funds. Modern portfolio theory (Markowitz, 1952) and stakeholder theory (Freeman, 1984) are set as the theoretical basis for this research as these theories provide competing propositions about the risk and RAR exposure of different types of funds.Two sampling approaches, both top sample and matched sample methods, are applied to identify the sample of funds that operate in the capital market of the United Kingdom. In the top sample, 50 ethical, 50 conventional and 15 Islamic funds that have the top asset sizes in each fund type are selected. In the matched sample, 50 conventional funds are matched for 50 top ethical funds and 11 ethical and 11 conventional funds are matched for the 11 top Islamic funds based on fund size, age, domicile, and investment area to create a sample of funds with similar characteristics. Data are collected for the sample period of July 2011 to June 2018.The risk and RAR measures are empirically compared using both univariate and multivariate data analysis techniques. In the univariate data analysis, both parametric and nonparametric tests are performed. In the multivariate analysis, the fund-specific characteristics are used as control variables to eliminate the effects of heterogeneous fund characteristics and fund style dimensions while comparing the risk and RAR measures across different fund types. Therefore, two types of samples and several data analysis techniques are employed in this study to produce robust analysis.The findings demonstrate that the risk exposure of ethical funds is neither entirely higher nor entirely lower than conventional funds; however, ethical funds are exposed to overall higher RAR than conventional funds. These findings suggest that ethical screening criteria do not always negatively impact on risk, but there is evidence of a positive impact on RAR. Furthermore, mixed results are identified regarding risk and RAR exposure across different ethical fund styles. In particular, there is evidence of both equal and different risk and RAR exposures across ethical fund styles.Furthermore, overall, the negative ethical fund (NEFs) style is exposed to lower or equal risk compared to other ethical fund styles, but the RAR of NEFs is higher or equal than other ethical fund styles. . Finally, the empirical evidence shows that in general Islamic funds are exposed to higher risk than both ethical and conventional funds; however, the RAR of Islamic funds in general does not significantly differ from the ethical and conventional funds. These findings indicate that stricter ethical screening criteria and Islamic faith-based screening do not affect the RAR of ethical funds styles and Islamic funds adversely
Pedagogical matters: a dialogue of diverse persons, perspectives, and programs
No full textObjectivesOur paper puts forward the contention that diversity of ideas will be essential if nurse education is going to have the flexibility, innovation, and dynamism needed to meet current and future challenges.MethodsWe describe a unique crucible of innovation that was forged through a weekly discussion of international nurse educators.ResultsOur paper sets out the resulting outcomes, stimulating new connections, curriculum development, studies, and shared learning.ConclusionsWe offer insights from participants’ assessment of the group’s effectiveness to drive innovation in nursing education. In so doing, we urge educators to reach out, to connect, and to use resulting opportunities to learn
Smart Grid Enabler (SGE) framework development for residential sector towards energy conservation strategies via Fuzzy Delphi method
Get PDFThe global implementation of smart grids has revolutionised the electricity grid, including in Malaysia, where Tenaga Nasional Berhad leads the initiative. This Smart Grid system is akin to the internet of electricity, serving as a pivotal element in the sustainable energy transition within the electricity sector. The residential sector, which heavily relies on electricity for daily activities, stands to benefit significantly from this development. Despite the numerous advantages offered by smart grids, the Malaysian residential sector has not yet fully optimised these benefits to incorporate fit-for-use components. Thus, this research aims to develop a Smart Grid Enabler (SGE) framework for the Malaysian residential sector that supports energy conservation strategies. Based on the Trias Energetica and Smart Grid Conceptual Model, the Fuzzy Delphi Method is utilised to synthesise expert consensus and establish critical components for the SGE framework and achieve the aim of this research. Subsequent future research anticipates a better refined SGE framework which explores pragmatic and practical real-world applications in the residential setting. This study contributes to ongoing efforts in sustainable energy development through smart grid system optimisation for the Malaysian residential sector
Data sources:
Access Repository Dashboard
Do you manage Repository@Napier? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!