Birmingham City University Open Access Repository
Not a member yet
8823 research outputs found
Sort by
Mindful Mindsets and Rural Community Characteristics in Promoting Sustainable Rural Tourism and Facilitating the Tangible Implementation of the Circular Economy
No full textPurpose—This study aims to examine the role of mindful consumption in promoting rural sustainability, particularly in the context of tourism in Muang Kaen Community, Chiang Mai, Thailand, by establishing a robust circular economy.
Design/methodology/approach—The data were collected through in-depth interviews with 28 informants who are tourism stakeholders regarding sustainable development, i.e., government officers, business owners, community leaders, and community members in Muang Kaen, to achieve the data triangulation. A thematic analysis of the interview data was employed in this data set.
Findings - The findings demonstrate three key themes for driving sustainable community development: a sense of community, leadership, and embodiment. At an individual level, local community members co-create a sense of community through Thainess, which gradually forms the social commitment to caring for neighbors, the community, and the environment. Carefulness also relates to another theme, ‘leadership’ – social capital, which drives mindful behavior among the community members. Both situational and official leaders are key persons in forming a culture of sustainability within the community. Finally, the community can achieve sustainable goals by driving from the individual to the collective level through the embodiment.
Research limitations/implications - This single-case study warrants further examination across different communities to generalize the findings to broader circumstances.
Originality/value - This study has shed light on how rural tourism can drive sustainable development through a Circular Economy and Mindful Consumption
Post weld heat treatment of bisalloy 80 steel: Mechanics and industry safety code compatibility
No full textThe present study utilized both in-situ and ex-situ neutron diffraction to evaluate the evolution of residual strains/stresses before, during and after Post-Weld Heat Treatment (PWHT) of quenched and tempered (Q&T) Bisalloy 80 steel welded by pulsed Gas Metal Arc Welding (GMAW-P). It was found that strain/stress relaxation mainly occurred during the reheating step with a high relaxation rate and steep slope (∼67 % of strain relaxation) while linear strain relief was observed during holding (soaking) time. Most of the strain relief occurred within the temperature range of 450°C–600 °C which is believed to be due to creep strain development occurring far earlier than the component reaching the isothermal holding temperature. The ex-situ neutron diffraction measurements were similar to in-situ results confirming the applied PWHT effectively mitigated the residual stresses (the maximum longitudinal stress reduced to around 23 % of the weld metal yield strength).
The measurements were compared with existing literature data and the current fitness of safety assessment codes (BS7910 and R6). It was found that both assessment codes were conservative for both the transverse and longitudinal residual stresses in the region close to the weld toe. Furthermore, both standards may underestimate through-thickness residual stresses in the transverse direction
Early Detection of Skin Diseases Across Diverse Skin Tones Using Hybrid Machine Learning and Deep Learning Models
No full textSkin diseases in melanin-rich skin often present diagnostic challenges due to the unique characteristics of darker skin tones, which can lead to misdiagnosis or delayed treatment. This disparity impacts millions within diverse communities, highlighting the need for accurate, AI-based diagnostic tools. In this paper, we investigated the performance of three machine learning methods—Support Vector Machines (SVMs), Random Forest (RF), and Decision Trees (DT)—combined with state-of-the-art (SOTA) deep learning models: EfficientNet, MobileNetV2, and DenseNet121, for predicting skin conditions using dermoscopic images from the HAM10000 dataset. Features were extracted using the deep learning models, with labels encoded numerically. To address data imbalance, SMOTE and resampling techniques were applied. Additionally, Principal Component Analysis (PCA) was used for feature reduction, and fine-tuning was performed to optimize the models. The results demonstrated that RF with DenseNet121 achieved superior accuracy of 98.32%, followed by SVM with MobileNetV2 at 98.08%, and Decision Tree with MobileNetV2 at 85.39%. The proposed methods overcome the SVM with SOTA EfficientNet model, validating the robustness of the proposed approaches. Evaluation metrics such as accuracy, precision, recall, and F1-score were used to benchmark performance, showcasing the potential of these methods in advancing skin disease diagnostics for diverse populations
Enhancing Facial Recognition Accuracy in eKYC Systems: A Comparative Evaluation of Euclidean Distance, Cosine Similarity, and SSIM Under Real-World Challenges
No full textEnsuring robust banking security is an ongoing challenge, especially in the face of increasingly sophisticated fraud techniques. One critical aspect is the accuracy of facial recognition technology, which plays a central role in elec-tronic Know Your Customer (eKYC) processes. Inaccurate facial recognition can result in security breaches, allowing fraudsters to exploit vulnerabilities during customer registration and transactions. This issue is particularly signif-icant in Thailand’s banking industry, where reliance on eKYC frameworks is growing. Current facial recognition methods often struggle with false posi-tives, impersonation, and spoofing attacks, threatening the integrity of finan-cial systems.
Hence, this paper addresses these concerns by exploring both the limitations of existing face recognition techniques and the opportunities presented by recent advancements in deep learning. The primary goal of this research is to enhance the accuracy of face recognition systems used in eKYC through the integration of advanced algorithms. By refining facial feature extraction methods and employing adaptive learning models, we aim to reinforce the verification process and significantly reduce the risk of fraud
AI-Powered Skin Disease Diagnosis Across Diverse Tones
No full textSkin diseases in melanin rich skin often present diagnostic challenges, as their detection is more difficult due to the characteristics of darker skin tones, leading to misdiagnosis or delayed treatment. This disparity affects millions of individuals in the Black community. Developing accurate and re-liable AI-based automated tools is crucial to address these challenges. In this project, we applied three machine learning models and a data mining ap-proach to accurately predict various skin conditions using dermoscopic im-ages. Employing the CRISP-DM methodology, we implemented three tech-niques: Convolutional Neural Networks (CNNs), known for their strength in image recognition; Support Vector Machines (SVMs), effective in distin-guishing subtle differences in skin conditions; and Decision Trees, a simpler yet interpretable model. These were applied to predict skin disease types us-ing the HAM10000 image dataset. To manage data imbalance, SMOTE and resampling techniques were employed. Additionally, PCA and fine-tuning were used to optimize the models. Accuracy is selected as evaluation met-rics. Among the models, SVM with the RBF kernel achieved the highest ac-curacy of 88.48%, followed by Decision Trees with 82.48% and CNN-ResNet50 with 77.28%. Comparisons with other HAM10000 dataset predic-tion methods demonstrated superior performance for the proposed approach, particularly with the SVM-RBF model. These results suggest that our meth-ods for skin disease detection have significant potential to benefit healthcare by offering more accurate and reliable diagnostic tools
From Thousands of African Languages to a Pan-African Language for the African Continental Free Trade Area: A Framework Promoting Kiswahili as Common Language for Intra-African Trade
No full textThis paper proposes a framework for promoting Kiswahili as the pan-African trade language to support the African Continental Free Trade Area (AfCFTA) in achieving its goal of boosting intra-African trade on a continent that has over 3,000 spoken languages and no common lingua franca. We draw from Scott’s (1995) institutional theory to structure the promotion of Kiswahili around the regulative, normative, and cognitive pillars. The framework aims to stimulate scholarly discussion on a pan-African trade language and serve as a tool for policymakers at national and supranational levels to promote Kiswahili as a pan-African language for trade
Advanced Diagnosis of Cardiac and Respiratory Diseases from Chest X-Ray Imagery Using Deep Learning Ensembles
No full textChest X-ray interpretation is essential for diagnosing cardiac and respiratory diseases. This study introduces a deep learning ensemble approach that integrates Convolutional Neural Networks (CNNs), including ResNet-152, VGG19, EfficientNet, and a Vision Transformer (ViT), to enhance diagnostic accuracy. Using the NIH Chest X-ray dataset, the methodology involved comprehensive preprocessing, data augmentation, and model optimization techniques to address challenges such as label imbalance and feature variability. Among the individual models, VGG19 exhibited strong performance with a Hamming Loss of 0.1335 and high accuracy in detecting Edema, while ViT excelled in classifying certain conditions like Hernia. Despite the strengths of individual models, the ensemble meta-model achieved the best overall performance, with a Hamming Loss of 0.1408 and consistently higher ROC-AUC values across multiple diseases, demonstrating its superior capability to handle complex classification tasks. This robust ensemble learning framework underscores its potential for reliable and precise disease detection, offering significant improvements over traditional methods. The findings highlight the value of integrating diverse model architectures to address the complexities of multi-label chest X-ray classification, providing a pathway for more accurate, scalable, and accessible diagnostic tools in clinical practice
Enhancing Resilience in IoT Water Systems Using Data-Intelligence and Decentralization
No full textIn recent years, concerns regarding the security of water networks have escalated due to the increasing integration of water assets (actuators and sensors) with the Internet, combining Information Technology (IT) and Operation Technology (OT). This integration promises improved services for water networks but also introduces the risk of cyber-attacks and physical threats. As a result, there is a growing need for novel security measures to protect integrated Cyber-Physical Systems (CPS) in water distribution systems (WDSs). This article assesses actual incidents and potential Cyber-Physical (CP) attacks on water systems, explores their operational impacts, and suggests mitigating measures. It introduces a secure architecture for an integrated CPS in WDS. The study incorporates attack detection and data validation models to enhance system robustness and reduce risks, adhering to the security criteria of Water 4.0. First, the attack detection model utilizes a two-stage architecture employing six Machine-Learning (ML) algorithms, resulting in developing a simulation model with the best-suited configuration. Second, the data validation model uses blockchain technology on transmitted data, creating a simulation model for water consumption data with various input types, consensus mechanisms, and data output conversion methods. Finally, this article provides a foundation for researchers, professionals, and operators in the water sector to experiment with, evaluate, and further develop this secure architecture for their water systems. Simulating their networks using the proposed architecture allows them to identify the most suitable configurations and parameters for their specific implementations
Creating a Teacher Educators Digital Hub (TEDH): A case study of digital transformation in two teacher training universities in Viet Nam
No full textThe key aims of the TEDH project were to design and create a digital hub of training and teaching materials and resources, across a range of disciplines, for teacher educators in Viet Nam. The hub includes a bespoke training ‘how to’ course on core digital skills, and prototype digital resources that can be downloaded and/or adapted by teacher educators. TEDH is also developing a ‘community support area’ so that resources and teaching ideas can be shared, or educators can ask questions and seek peer support. In addition, teacher educators have access to a coaching and mentoring programme to support them using the hub. The development of the hub and the resources have been informed by evaluations with teacher educators