Great Britain Journals Press
Not a member yet
951 research outputs found
Sort by
Classification of Acoustic Data with Transformer Model
Full text linkBees are essential to global ecosystems, particularly for pollinating crops, yet in recent years their populations have faced significant decline. One critical aspect of bee colony health is the ability to detect negative in-hive events such as a queen leaving the hive. Traditionally, beekeepers rely on manual inspections to assess hive conditions, a labor-intensive and time-consuming process. However, recent advances in machine learning offer new approaches to automating this task. Since 2016, there have been attempts to classify bee sounds using machine learning, employing the power of different machine learning methods, including deep learning architectures.
In this research, we explore the use of acoustic labeled data for in-hive event classification, focusing specifically on detecting when a queen leaves the hive. We utilize 12-hour recordings from different locations, with the data preprocessed and transformed to be suitable for input into a transformer-based neural network. Our goal is to demonstrate that transformer models yield superior results in this task compared to previous approaches. The study is organized into several key sections: we first highlight the ecological importance of bees, followed by a literature review on the state of bee sound classification research. We then delve into the data preparation process, model design, and present our findings. Our results underscore the potential of transformer models in automating hive monitoring, offering a scalable solution for beekeepers to protect and preserve bee populations
Enhancing Pairs Trading Strategies in the Cryptocurrency Industry using Machine Learning Clustering Algorithms
Full text linkConventional pair trading methods, which rely on statistical and linear assumptions, often struggle to cope with the high volatility and dynamic nature of cryptocurrency markets. This study explores how pair trading strategies might be improved by using machine learning clustering algorithms to uncover latent links between cryptocurrencies. Specifically, it employs unsupervised clustering techniques k-means, hierarchical clustering, and affinity propagation on daily closing prices of the top 50 cryptocurrencies, selected based on their market capitalization and daily trading volume, from January 2021 to November 2024. The methodology includes data preprocessing, exploratory data analysis, clustering, and cointegration tests for pair selection. The main findings show that clustering algorithms can efficiently group cryptocurrencies based on similar behavioural price patterns, with affinity propagation outperforming other models in cluster definition. The study reveals 21 pairs with strong cointegration strategies among the chosen cryptocurrencies, indicating their appropriateness for trading. The study highlights the effectiveness of clustering algorithms in tackling cryptocurrency market volatility, optimizing pair selection, and adapting to dynamic conditions. It emphasizes the transformative potential of machine learning in enhancing trading techniques and efficiency in the cryptocurrencies market. The practical implications include advancing trading strategies for cryptocurrencies investors by incorporating machine learning algorithms to enhance market efficiency and profitabilit
Impact of Managerial Overconfidence, Internal Control and Cash Holdings in Chinese Firms
Full text linkThis study systematically evaluates the effects of managerial overconfidence on corporate capital holdings. According to the theoretical framework, this study conducted an empirical analysis using data on listed Chinese companies from 2010 to 2022. Traditional economic theories are based on rational human assumptions, and theories that explain corporate cash holdings concentrate on the potential uses of corporate cash assets; however, they neglect to consider decision-makers� beliefs that ultimately dictate cash utilization and cannot verify the difference between managers overconfidence and over-optimism. Therefore, this study not only verify the difference between managers overconfidence and over-optimism, but also based on psychology theory to explain why cognitive bias leads managers to be more overconfident in China and affects corporate cash holding levels. These findings indicate that managers� overconfidence positively associated with corporate cash holdings. Additionally, the high-quality internal controls mitigate overconfident managers holding more cash by alleviating information asymmetry. Collectively, this study effectively connects cognitive psychology to enrich the research on the correlation between managerial overconfidence and cash holdings. In addition, these findings illustrate the important implications of enforcing cash-use efficiency and alleviating cognitive bias in overconfident managers
The Impact of Gross Domestic Product on Co2 Emissions (A Case Study of Asian Tiger Countries)
Full text linkThis research examines the impact of Gross Domestic Product (GDP) on CO2 emissions in four well-developed Asian countries�South Korea, Singapore, Taiwan, and Hong Kong�over the period from 1960 to 2019. To analyze the relationships, regression was performed using the Generalized Least Squares (GLS) method in STATA-14. The results indicate that all regressors are significant, and to address the issue of autocorrelation in the model, an Autoregressive Lag model was used. By adding the lag of an independent variable to the model, the problem of autocorrelation was resolved. Consequently, the model\u27s goodness-of-fit improved, and the significance levels of the regressors were confirmed. Based on the research findings, it can be concluded that the economic growth of these countries leads to an increase in carbon dioxide emissions into the external environment
Blockchain and AI in Reverse Logistics: A Qualitative Synthesis of Strategic Applications and Challenges
Full text linkSince the rapid growth of e-commerce, product returns volume has soared, creating� pressure on reverse logistics systems to be more efficient, transparent and cost-effective.� Nonetheless, the traditional reverse logistics processes are encountered with operational� challenges such as high costs, return fraud, ineffective tracking systems, and� environmental concerns. Mature AI and Blockchain Technologies allow businesses to� automate various processes, make more informed decisions, and ensure that supply chain� operations are transparent.�
Based on case studies from leading global companies including Amazon, Walmart, and� Alibaba, this study sheds light on the integration of AI and Blockchain in reverse� logistics as an avenue for reverse logistics integration in overall supply chain strategy.� Employing a qualitative methodology and secondary data sources, including industry� reports, academic literature, and company publications, this study explores how these� modern technologies are positively impacting return management processing, fraud� detection, operational efficiency, and sustainability.�
AI-powered automation greatly helps the companies in returns forecasting, inventory� optimization, and customer service, unveiling the time and costs relating to reverse� logistics operations as per the findings. Simultaneously, the implementation of� blockchain technology allows for real-time tracking, fraud mitigation, and generation of� trusted data for sharing, thereby enhancing return transactional transparency.� Specifically, companies which combine both AI and blockchain, directly creates use� cases leading to superior decision-making capabilities, more efficient logistics and� support processes, as well as superior management of product returns.��
While these technologies offer potential advantages, the study reveals critical hurdles in� implementing these technologies, such as high deployment costs, regulatory obstacles,� data privacy issues, and interoperability challenges among various blockchain systems.� Moreover, it can also be quite challenging for companies to find the technical expertise� needed to implement AI or blockchain in their logistics infrastructure.
To the best of our knowledge, this research effort contributes to the literature on supply� chain management and technology adoption by providing tested instruments and best� practice options for businesses that seek to enhance their reverse logistics operations� through the application of technologies. The study also aids theory and practice by� illustrating some policies for adoption challenges and offering managerial� recommendations for firms. These insights need to be validated and refined through� empirical research and industry application of AI and blockchain in the context of� reverse logistics (Wang et al., 2018; Kafeel et al., 2023).
A Framework for Dynamic ANN Index Lifecycle Management in Ad Retrieval
Full text linkThe digital advertising landscape requires exceptional scale and efficiency in candidate retrieval systems, where Approximate Nearest Neighbor indexes function as the core technology facilitating real-time ad matching across extensive inventories. Dynamic advertising catalogs face distinct challenges due to ongoing changes from campaign launches, budget modifications, and creative updates, requiring ANN indexes that can manage frequent insertions, deletions, and changes while ensuring optimal query performance. Current dynamic ANN implementations experience a gradual performance decline as update operations build up, resulting in heightened query latency and diminished retrieval accuracy that directly affects system efficiency. The suggested framework tackles these issues using a thorough method that merges smart index selection techniques with advanced lifecycle management strategies. The selection element assesses candidate ANN algorithms based on workload-specific traits, including degradation resistance as a key factor in addition to conventional static performance measurements. The management part employs a dual-layer architecture that differentiates real-time updates from batch optimizations, allowing for instant responsiveness while maintaining long-term performance traits. At the core of this framework is a re-indexing policy that is aware of degradation, which observes performance indicators and initiates reconstruction actions using predictive models and adjustable thresholds. Experimental validation shows the framework\u27s efficacy in various scenarios that reflect production advertising environments, sustaining steady performance over long operational durations, whereas traditional methods need regular manual input. The framework allows for consistently high recall and minimal latency during millions of update operations, greatly surpassing conventional update-and-ignore methods typically used in dynamic indexing systems
Effect of Security on the Performance of Momo Data Promoter Agents at MTN Ivory Coast
Full text linkThis study aims to highlight the role of security in the performance of MTN Ivory Coast promoter agents. It focuses on a sample of three MTN network promoters residing in the municipalities of Koumassi, Adjam� and Yopougon. These promoters were subjected to the interview technique. The data from this operation were analyzed and followed by a synthesis. This reveals to us that security at the level of promoters and technical services play a key role in the performance of Momo Data promoters at MTN. This study recommands that managers of all compagnies take into account individual and IT security in oder to achieve satisfactory results
Artificial Intelligence and Counselling in Nigerian Schools: Confronting Challenges while Unlocking Prospects
Full text linkCounselling in Nigerian schools remains constrained by limited personnel, inadequate training, and infrastructural challenges, leaving many students underserved in their academic, psychological, and socio-emotional needs. With counsellor�student ratios far exceeding recommended standards, traditional approaches struggle to meet the demands of Nigeria�s rapidly expanding school population. This paper explore Artificial Intelligence and Counselling in Nigerian Schools: Confronting Challenges while Unlocking Prospects.� Artificial Intelligence (AI) offers promising solutions through tools such as chatbots, predictive analytics, and intelligent tutoring systems, which enhance accessibility, efficiency, and early identification of at-risk students. Drawing on the Diffusion of Innovations (DOI) framework, this paper examines both the opportunities and constraints of integrating AI into Nigerian school counselling, including issues of cultural localisation, ethical governance, infrastructural readiness, and sustainability. The discussion highlights hybrid models that combine AI-driven interventions with human expertise as the most viable pathway, ensuring empathy, ethical judgment, and cultural sensitivity are preserved. By outlining benefits, risks, and implementation strategies, the paper argues that responsible AI adoption can transform Nigeria�s counselling ecosystem, bridging gaps in student welfare and promoting equitable, inclusive educational outcomes
Study of Epidemiological Aspects of Cholera in the Edea Health District, Littoral–Cameroon
Full text linkIntroduction: Cholera is an acute diarrhoeal disease caused by infection of the intestine with Vibrio cholerae type O1 or O139 bacteria, which can lead to rapid dehydration and death. Both children and adults can be infected. Cholera is closely associated with poverty, poor sanitation and the absence of safe drinking water. As a result, the burden of cholera is concentrated in Africa and South Asia, accounting for around 99% of cases worldwide. Against a backdrop of water shortages in the town of Edea, and following confirmation of the existence of an outbreak of cholera in the town, a number of investigations were carried out to describe the event in terms of time, place and people, and to put control measures in place. The aim of this article is to report on the cholera epidemic observed in the Edea health district with a view to strengthening disease surveillance.
Methods: We conducted a descriptive cross-sectional study from the 1st to the 52nd epidemiological week of 2022 in the Edea health district, Littoral region, Cameroon. A linear list was drawn up, and cases were actively sought in the consultation registers of the health facilities and in the community. All suspected cholera cases notified in the health areas of the Edea health district were included in this study. Demographic, clinical, origin and outcome variables were extracted from the linear cholera list of the Edea Health District and analysed using EasyMedStat (version 3.22) and Microsoft Excel 2016.
Results: A total of 34 suspected cholera cases were identified in the district, including 2 deaths (5.88%). Of the suspected cases, 10 were RDT positive and 1 was confirmed by culture (V. cholerae O1 or O139); all over a period of 52 epidemiological weeks in 2022. The sex ratio F/M was 1.61. All age groups were affected. The most common age group was ?15 years, with 28 cases (82.3%). Most of the cases came from the Malimba health area: 9 cases (26.4%). All cases presented with diarrhoea and vomiting. Moderate dehydration was most frequently observed: 14 cases (41%).
Conclusion: A cholera epidemic has been confirmed in the Edea Health District. In order to contain future epidemics and move towards the elimination of the disease, it is essential to strengthen epidemiological surveillance and the application of preventive measures against cholera in the district, particularly during the rainy season, and to promote multi-sectoral collaboration through the involvement of all stakeholders in related sectors
Primary Pulmonary Resection in Madagascar: Indication and Results
Full text linkIntroduction:� Primary pulmonary resection is defined as the surgical ablating or endoscopic of an entire lobe called lobectomy or an entire lung called pneumonectomy. Our objective is to describe the indication for a primary pulmonary resection and to research the morbidity and mortality factors of primary pulmonary resection at the Joseph Ravoahangy Andrianavalona Antananarivo University Hospital.�
Method: This is a retrospective descriptive and analytical study of 216 patients hospitalized in the thoracic surgery department at the Joseph Ravoahangy Andrianavalona Antananarivo University Hospital (CHU-JRA), from January 1, 2015, to December 31, 2023, who underwent primary pulmonary resection.�
Results:�We collected 216 patients with a male predominance (73,6%) and a median age of 33. Post-tuberculosis pulmonary destruction is the main indication for resection in 53,24% cases, followed by a cavernous lesion of pulmonary aspergillosis in 24,07% cases then tumor mass in 18,06% and the nodular lesion in 4,63% cases. We realized a lobectomy in 71,29% of cases and a pneumonectomy in 28,71%. The complications found are dominated by pneumothorax, prolonged bubbling, pleural empyema, bronchopleural fistula, bleeding, recurrent paralysis, septic shock and cardiac rhythm disorder. The mortality rate is 8,33% cases. severe factors were found notably: undernutrition with IMC ? 18 kg/m2 (RR= 5[2,1-11,7]), hemorrhagic shock (RR=9,3[3,7-13,3]), septic shock (RR=13,3 [4,6-28,4]), cardiogenic shock (RR=8,5[3,7-12,3]), pleural empyema� (RR=8,5[3,8 -13,6], bronchopleural fistula (RR=6,7[2,9-15,3]).