ResearchBerg
Not a member yet
161 research outputs found
Sort by
Impact of Augmented Reality on Purchase Intention of Foreign Products Online
Get PDFAugmented reality (AR) is a significant technology that holds the promise to transform how consumers interact with products before purchasing. It creates immersive experiences that enable people to engage with digital material in a more intuitive and straightforward manner. When used effectively, AR can be influential in every stage of customer journey including purchase intention stage. Assessing purchase intention of international consumers is critical for organizations because it allows them to plan and make choices about marketing, inventory, and expenses. Purchase intent provides international companies with information on what their global consumers are willing to purchase enabling them to modify their marketing and goods to better fit their customers\u27 demands. This research examined how augmented reality increase the purchase intention of global customer using the data, which includes data for 810 different overseas visitors of an e-commerce site. We collected these data from visitors of a global e-commerce shop that integrated augmented reality (AR) into their smartphone app to enable users to imagine how they would appear with various items. The study performed a Robust Least Squares Method-estimation. Our research\u27s findings provide some early proof that using AR increases the level of purchase intention of foreign products. The findings also indicate that price, and the number of positive reviews increase the purchase intention of foreign products. Customers\u27 buying intentions may help firms predict future trends and organize their strategy appropriately. Businesses must also understand the elements that drive purchase intent, such as immersive experience with AR, consumer demographics, nationality, product attributes, pricing, and customer experience.
 
Smart Cities, Healthy Citizens: Integrating Urban Public Health in Urban Planning
Get PDFUrban planning that incorporates public health considerations is crucial for the development of smart cities that prioritize the well-being and health of their citizens. This study presents key findings on integrating urban public health into urban planning to create environments that promote physical and mental well-being. The study identifies and explores several crucial considerations for achieving this integration.The first consideration is healthy urban design, which involves designing urban spaces and infrastructure that promote physical activity, accessibility, and safety. Walkable neighborhoods, well-connected sidewalks, bike lanes, and efficient public transit systems encourage active transportation. Incorporating parks, green spaces, and recreational facilities provide opportunities for exercise and outdoor activities, while inclusive and accessible public spaces reduce pollution and noise.Air quality and pollution control emerge as another vital consideration. The study highlights the importance of implementing policies to mitigate air pollution, reduce emissions, and promote clean energy sources. Designing urban areas to minimize exposure to pollution sources, increasing green spaces and urban forests, and utilizing smart technologies for monitoring air quality are key strategies for improving air quality and mitigating the heat island effect.Ensuring accessible healthcare and services is essential for equitable public health. The research emphasizes the need to strategically locate healthcare facilities to serve both urban and underserved areas. Attention should be given to the needs of vulnerable populations, such as the elderly, low-income communities, and individuals with disabilities. The integration of telemedicine and digital health solutions can enhance access to healthcare services.Promoting active transportation and safety is crucial in urban planning. The study highlights the importance of pedestrian and cyclist safety through well-designed crosswalks, traffic calming measures, and lighting systems. Dedicated cycling infrastructure, traffic management strategies, and smart traffic systems contribute to reducing accidents and improving road safety.Noise pollution management is an often overlooked aspect of urban planning. The research emphasizes the significance of designing buildings with sound insulation and implementing zoning regulations to separate noise-sensitive areas from noise-generating activities. Green buffers and sound barriers are effective in mitigating noise impacts, while monitoring noise levels and enforcing regulations minimize excessive noise.The study also underscores the importance of integrating elements that promote mental health and social well-being into urban planning. Creating inclusive and socially connected neighborhoods, designing public spaces that encourage socialization and relaxation, and prioritizing the provision of community centers and social services all contribute to mental health and well-being.Data and technology integration play a crucial role in informing urban planning decisions and improving public health outcomes. The study highlights the value of collecting and analyzing health-related data to identify health disparities, understand the impact of the built environment on health, and guide decision-making processes. Utilizing smart technologies, such as wearable devices and health monitoring systems, promotes individual health awareness and facilitates targeted interventions.Evaluation and monitoring are essential components of successful urban planning. Continuously monitoring and evaluating the impact of urban planning decisions on public health outcomes, collecting data on health indicators, and using this information to assess intervention effectiveness and inform future planning efforts are critical for sustainable development.Integrating urban public health considerations into urban planning enables the creation of smart and healthy environments that support the well-being of citizens. This holistic approach ensures that urban development fosters economic growth, technological advancement, and the health and happiness of the people who live and work in these cities
Pharmacovigilance in Developing Countries: Drivers and Barriers
Get PDFThe Pharmacovigilance is crucial for patient safety and improving healthcare quality, particularly in developing countries where the burden of disease is high. This study examines the drivers and barriers to implementing effective pharmacovigilance systems in these countries. Increased availability and use of medicines and the globalization of the pharmaceutical industry have led to an increase in the number of adverse drug reactions and a need for harmonized pharmacovigilance regulations and reporting standards. Developing countries have made significant progress in establishing regulatory frameworks, and international organizations have provided technical assistance and resources. The use of technology, such as mobile phone applications, has made it easier to report adverse drug reactions and monitor drug safety in developing countries. However, several barriers hinder the success of pharmacovigilance systems in developing countries. These include the lack of resources, limited awareness and education, and limited access to information, making it challenging for healthcare professionals and regulatory authorities to monitor drug safety. Fragmented healthcare systems and cultural and social barriers, such as stigma associated with reporting ADRs, further compound the issue. It is essential to address these barriers to promote the successful implementation of pharmacovigilance systems in developing countries. Doing so will improve patient safety, reduce the burden of disease, and enhance healthcare quality. Future research should focus on developing strategies to overcome these barriers and promote the effective implementation of pharmacovigilance systems in developing countries
Analysis of Cloud Based Keystroke Dynamics for Behavioral Biometrics Using Multiclass Machine Learning
Get PDFWith the rapid proliferation of interconnected devices and the exponential growth of data stored in the cloud, the potential attack surface for cybercriminals expands significantly. Behavioral biometrics provide an additional layer of security by enabling continuous authentication and real-time monitoring. Its continuous and dynamic nature offers enhanced security, as it analyzes an individual\u27s unique behavioral patterns in real-time. In this study, we utilized a dataset consisting of 90 users\u27 attempts to type the 11-character string \u27Exponential\u27 eight times. Each attempt was recorded in the cloud with timestamps for key press and release events, aligned with the initial key press. The objective was to explore the potential of keystroke dynamics for user authentication. Various features were extracted from the dataset, categorized into tiers. Tier-0 features included key-press time and key-release time, while Tier-1 derived features encompassed durations, latencies, and digraphs. Additionally, Tier-2 statistical measures such as maximum, minimum, and mean values were calculated. The performance of three popular multiclass machine learning models, namely Decision Tree, Multi-layer Perceptron, and LightGBM, was evaluated using these features. The results indicated that incorporating Tier-1 and Tier-2 features significantly improved the models\u27 performance compared to relying solely on Tier-0 features. The inclusion of Tier-1 and Tier-2 features allows the models to capture more nuanced patterns and relationships in the keystroke data. While Decision Trees provide a baseline, Multi-layer Perceptron and LightGBM outperform them by effectively capturing complex relationships. Particularly, LightGBM excels in leveraging information from all features, resulting in the highest level of explanatory power and prediction accuracy. This highlights the importance of capturing both local and higher-level patterns in keystroke data to accurately authenticate users
Harnessing Machine Learning to Improve Healthcare Monitoring with FAERS
Get PDFThis research study investigates the potential of machine learning techniques to improve healthcare monitoring through the utilization of data from the FDA Adverse Event Reporting System (FAERS). The objective is to explore specific applications of machine learning in healthcare monitoring with FAERS and highlight their findings. The study reveals several significant ways in which machine learning can contribute to enhancing healthcare monitoring using FAERS.Machine learning algorithms can detect potential safety signals at an early stage by analyzing FAERS data. By employing anomaly detection and temporal pattern analysis techniques, these models can identify emerging safety concerns that were previously unknown or underreported. This early detection enables timely action to mitigate risks associated with medications or medical products.Machine learning models can assist in pharmacovigilance triage, addressing the challenge posed by the large number of adverse event reports within FAERS. By developing ranking and classification models, adverse events can be prioritized based on severity, novelty, or potential impact. This automation of the triage process enables pharmacovigilance teams to efficiently identify and investigate critical safety concerns.Machine learning models can automate the classification and coding of adverse events, which are often present in unstructured text within FAERS reports. Through the application of Natural Language Processing (NLP) techniques, such as named entity recognition and text classification, relevant information can be extracted, enhancing the efficiency and accuracy of adverse event coding.Machine learning algorithms can refine and validate signals generated from FAERS data by incorporating additional data sources, such as electronic health records, social media, or clinical trials data. This integration provides a more comprehensive understanding of potential risks and helps filter out false positives, facilitating the identification of signals requiring further investigation.Machine learning enables real-time surveillance of FAERS data, allowing for the identification of safety concerns as they occur. Continuous monitoring and real-time analysis of incoming reports enable machine learning models to trigger alerts or notifications to relevant stakeholders, promoting timely intervention to minimize patient harm.The study demonstrates the use of machine learning models to conduct comparative safety analyses by combining FAERS data with other healthcare databases. These models assist in identifying safety differences between medications, patient populations, or dosing regimens, enabling healthcare providers and regulators to make informed decisions regarding treatment choices.While machine learning is a powerful tool in healthcare monitoring, its implementation should be complemented by human expertise and domain knowledge. The interpretation and validation of results generated by machine learning models necessitate the involvement of healthcare professionals and pharmacovigilance experts to ensure accurate and meaningful insights.This research study illustrates the diverse applications of machine learning in improving healthcare monitoring using FAERS data. The findings highlight the potential of machine learning in early safety signal detection, pharmacovigilance triage, adverse event classification and coding, signal refinement and validation, real-time surveillance and alerting, and comparative safety analysis. The study emphasizes the importance of combining machine learning with human expertise to achieve effective and reliable healthcare monitoring
Cloud Security and Risk Prevention with Artificial Intelligence: Designing Effective Anomaly Detection Frameworks for Distributed Architectures
Get PDFThe adoption of cloud computing has led to an exponential growth in data storage and processing capabilities, enabling businesses to achieve unprecedented scalability and operational efficiency. However, the distributed nature of cloud environments introduces significant security risks, including data breaches, unauthorized access, and system compromises. Traditional security mechanisms often fall short in addressing these dynamic threats due to the complexity and scale of cloud architectures. Artificial intelligence (AI), particularly anomaly detection frameworks, has emerged as a pivotal tool in cloud security by enabling real-time monitoring, threat identification, and adaptive risk prevention. This paper explores the integration of AI-driven anomaly detection systems within distributed cloud architectures, emphasizing their design, implementation, and efficacy in mitigating security threats. We discuss key methodologies, including supervised, unsupervised, and hybrid learning techniques, for anomaly detection. Additionally, we analyze the challenges associated with distributed systems, such as latency, scalability, and false positives, and propose strategies to overcome them. This research also examines case studies where AI-based frameworks significantly improved the security posture of cloud systems. By leveraging advanced AI models, such as deep learning and reinforcement learning, this study demonstrates how adaptive anomaly detection frameworks can proactively address emerging threats in real-time. Ultimately, the findings underscore the importance of designing robust AI-driven frameworks to safeguard cloud infrastructures while minimizing operational disruptions
Quantifying Healthcare Consumers\u27 Perspectives: An Empirical Study of the Drivers and Barriers to Adopting Generative AI in Personalized Healthcare
Get PDFThis study empirically examines the drivers and barriers of adoption of Generative AI (GenAI) in personalized healthcare from the perspective of healthcare consumers. A quantitative analysis was conducted on a sample of 376 healthcare consumers, defined broadly to include individuals and their families or caregivers who interact with healthcare services. The research employed machine learning multiclass classification methods, correlation analysis, and a probabilistic ordered regression model. The data set consisted of 376 observations with 15 features, which was preprocessed, imputed for missing values, and analyzed using StratifiedKFold with 10 folds. The performance of each model was evaluated based on accuracy, Area Under the Curve (AUC), recall, precision, F1 score, Kappa, Matthews Correlation Coefficient (MCC), and execution time. The results showed that Linear discriminant analysis (LDA) emerged as the top-performing model with an accuracy of 95.09%, AUC of 99.68%, and an execution time of only 0.038 seconds. The study also revealed significant correlations between variables and GenAI adoption, highlighting Digital Divide as the most influential factor with a negative correlation of -0.123504. Feature importance analysis from the LDA model indicated that Understanding and Trust in AI Systems, Convenience and Accessibility, and Improved Health Outcomes were among the top influencers. In Logistic Regression and Random Forest models provided Digital Divide consistently appear as a significant barrier to adopt GenAI based healthcare services. A probabilistic ordered regression analysis further elucidated the impact of these variables on adoption willingness. It showed that approximately 89.26% of the variability in willingness to adopt GenAI was explained by the independent variables, with Digital Divide, Convenience and Accessibility, and Understanding and Trust in AI Systems being statistically significant. The overall findings of the study show that the adoption of GenAI in personalized healthcare is primarily driven by its potential to improve health outcomes, increase accessibility and convenience, and provide personalized care. Barriers such as data privacy concerns, trust issues, and digital divide must be addressed to facilitate wider adoption. The study highlights the need for healthcare providers and policymakers to focus on these key areas to enhance the acceptance and effectiveness of GenAI-based healthcare services
Forecasting Resource Usage in Cloud Environments Using Temporal Convolutional Networks
Get PDFBackground: Predicting resource usage in cloud environments is crucial for optimizing costs. While recurrent neural networks and time series techniques are commonly used for forecasting, their limitations, such as vanishing gradients and lack of memory retention, necessitate the use of convolutional networks for modeling sequential data.
Objective: This research proposes a temporal convolutional network (TCN) to forecast CPU usage and memory consumption in cloud environments. TCNs utilize dilated convolutions to capture temporal dependencies and maintain a fixed-sized receptive field, enabling them to handle sequences of varying lengths and capture long-term dependencies. The performance of the TCN is compared with Long Short-Term Memory (LSTM) Networks, Gated Recurrent Unit (GRU) Networks, and Multilayer Perceptron (MLP).
Dataset: The study employs the Google Cluster Workload Traces 2019 data, focusing on CPU and memory utilization ranging between 5% and 95% over a 24-hour period, extracted from the first ten days.
Results: The TCN outperforms other methods in predicting both CPU usage and memory consumption. For CPU usage prediction, the TCN achieves lower error metrics, including Mean Squared Error (MSE) of 0.05, Root Mean Squared Error (RMSE) of 0.22, Mean Absolute Error (MAE) of 0.18, and Mean Absolute Percentage Error (MAPE) of 3.5%. The TCN also demonstrates higher forecast accuracy, with FA1 = 85%, FA5 = 95%, and FA10 = 98%. Similar performance improvements are observed for memory consumption prediction, with the TCN achieving lower error metrics and higher forecast accuracy compared to LSTM, GRU, and MLP. The TCN exhibits better computational efficiency in terms of training time, inference time, and memory usage.
Conclusion: The proposed temporal convolutional network (TCN) demonstrates good performance in forecasting CPU usage and memory consumption in cloud environments compared to LSTM, GRU, and MLP. Since TCN\u27s can capture temporal dependencies and handle sequences of varying lengths makes it a promising approach for resource usage prediction and cost optimization in cloud computing
Impact of Diabetes and Hypertension Control on Work Performance among Employees in Malaysia
Get PDFDiabetes and hypertension are significant public health issues in Malaysia. Despite the abundance of treatments, the incidence of diabetes hypertension has grown. Working-age people are the most impacted by these and their associated complications, which have a negative impact on their quality of life and job productivity. Given the growing significance of diabetes and hypertension control and the sophistication of assessing its productivity losses, we present a critical evaluation of the roles that diabetes and hypertension control play in boosting employee performance. The empirical findings of this research showed that the work performance of employees significantly increases when diabetes and hypertension are controlled. Our study emphasizes the importance of prevention, treatment, and proper control of diabetes and hypertension among employees in Malaysia
The Economical Aspects of 3D Printing in Medical Practice: Cost-Benefit Analysis Across Cardiology, Neurosurgery, Imaging, and Urology
Get PDF3D printing technology has permeated the medical sector, promising transformative patient-specific interventions, ranging from surgical planning tools to tailored implants. This study undertook a comprehensive cost-benefit analysis to discern the economic viability of 3D printing in four distinct medical disciplines: cardiology, neurosurgery, imaging, and urology. Within cardiology, 3D models of patients\u27 hearts, primed for surgical planning, showed potential in decreasing operation durations and associated complications. Although offset by initial printer costs and requisite training, these reductions suggest a promising long-term financial advantage. Neurosurgical applications revealed similar benefits, especially for intricate surgeries where tangible models can reduce uncertainties. Considering the high stakes in neurological procedures, where complications can incur exorbitant costs, the preliminary investment in 3D printing emerges as justifiable. The realm of medical imaging, too, witnessed pronounced benefits. Translating MRIs, CT scans, and ultrasounds into physical models not only enhanced diagnostic precision but also proved instrumental in patient-education, elucidating complex conditions. Yet, this is balanced by the ongoing material costs and image conversion software. Lastly, in urology, the advantages of 3D printing appear more situation-specific. While there\u27s evident value in surgical planning for intricate kidney surgeries and crafting custom stents, widespread acceptance mandates unequivocal evidence of superior patient outcomes. The economic costs of introducing 3D printing in medical settings, though substantial, are often outweighed by its myriad benefits. As the technology matures and becomes more affordable, its adoption may present an even more favorable economic profile. Beyond tangible economic savings, the intangible benefits, such as amplified patient satisfaction and heightened surgeon assurance, further underscore the technology\u27s potential worth