As family-based care declines, community-based home help services have become increasingly important for older adults. This study explores how frail older adults who have lost their only child perceive and experience home help services in relation to their care needs. A descriptive phenomenological approach was adopted, using in-depth interviews with 11 participants in Chongqing, China, and thematic analysis. The findings show that home help services provide essential everyday support but are often insufficient to meet continuous, complex, or urgent care needs. Participants responded by adjusting their help-seeking practices and care expectations while balancing available support with personal needs. These findings highlight the gap between service provision and lived care needs, and suggest that improving the coordination and responsiveness of community-based home help services may better support this vulnerable population.
This study develops and empirically validates a conceptual model explaining how inspection activities influence self-assessment outcomes in higher education institutions, with particular emphasis on the mediating role of feedback quality and the moderating role of quality culture. A quantitative design using PLS-SEM was employed on a dataset of 377 respondents directly involved in quality assurance processes across Vietnamese higher education institutions. Inspection was operationalized as a multidimensional construct, while self-assessment outcomes were measured across three domains. Moderation and moderated mediation were further examined using regression-based approaches. The results reveal a highly asymmetric causal structure. Feedback quality emerges as a dominant amplification mechanism, exerting substantially stronger effects on outcomes than inspection variables. Mediation is selective rather than uniform, with only specific inspection dimensions operating through feedback. Contrary to theoretical expectations, quality culture does not exhibit significant moderating effects, indicating that its influence is structural rather than conditional. This study contributes to the literature by (i) reconceptualizing the inspection–feedback–improvement relationship as a mechanistic system, (ii) identifying feedback as an amplification mechanism rather than a passive mediator, and (iii) challenging the conventional treatment of quality culture as a moderating variable. The findings provide both theoretical refinement and practical implications for transitioning toward feedback-centered, learning-oriented quality assurance systems aligned with ESG 2015.
Deep neural networks are commonly trained using adaptive optimization methods such as Adam because they converge quickly and perform well under stochastic training conditions. Despite these advantages, recent research has revealed several important drawbacks of Adam. In particular, the optimizer tends to converge toward sharp minima, exhibits oscillatory behavior during loss optimization, and can produce unstable parameter updates when training in highly noisy environments. In this work, we propose NeuroFuzzyAdam (NF Adam), a novel optimizer that integrates fuzzy logic-based scalar modulation directly into the Adam update rule to improve learning stability and generalization. By introducing a bounded correction term via hyperbolic tangent transformations, NF Adam adaptively regulates step sizes based on moment estimates while preserving Adam’s key strengths. We provide theoretical convergence guarantees under convexity and bounded-gradient assumptions, matching Adam’s O(1/T) regret bound. Empirically, we evaluate the NF Adam across both image (MNIST, Fashion-MNIST, CIFAR-10) and tabular (UCI Iris, Wine, Breast Cancer) datasets using consistent architectures. Results show that NF Adam improves training smoothness, reduces gradient norm variance, and achieves competitive classification accuracy compared to Adam, SGD, RMSprop, and AdamW. These findings suggest that incorporating fuzzy logic into optimization dynamics presents a promising direction for enhancing robustness in neural network training, especially in scenarios with high gradient noise or non-smooth loss surfaces.
During the present period of cultural integration worldwide, there’s a heredity and innovation conundrum for traditional ethnic vocal arts. The current evaluation system is obviously too subjective and lacks a clear road. There is no clear direction for development. This paper uses an improved BP neural network and LSTM (Long-Short-Term-Memory Network) algorithm to construct a whole evaluation model. To accurately understand the effectiveness and innovativeness of traditional ethnic vocal art inheritance, and to transform the qualitative indicator values into feature vector sets through quantization and to input them into the neural network system so as to achieve an accurate judgment of the status quo of its development and a forecast of the future trends. Collecting multiple ethnic group data and constructing a model database based on traditional multiple school vocal databases. This is compared and analyzed with single-layer BP neural network and classic indicators. From the experimental test results, it can be seen that compared with the single architecture, the proposed integrated architecture has the advantages of faster convergence and higher prediction accuracy. And it also greatly reduces the data bias caused by human factors, which provides scientific basis for the spread and development of ethnic vocal cultural. It also expands the application area of technology for the digital protection of intangible cultural heritage.
This paper aims to make some contributions to the stability problems associated with linear neutral systems comprising multiple time delays in the states together with involving multiple neutral delays in time derivatives of states of system. We first develop an appropriate Lyapunov functional, and then derive new global asymptotic stability conditions for such a model of systems with the help of this functional. The proposed stability results basically depend on the constraint conditions that are imposed on the components of constant system matrices without involving delay parameters. It is also shown that the obtained stability conditions improve some formerly published corresponding stability conditions for the same model of neutral systems. This work also studies a comparative numerical example to indicate advantage of the results of this paper over the past results and demonstrate the applicability of the derived stability criteria of the algebraic forms.
The purpose of this research is to determine whether Indonesian millennials' experiences and their lifestyle regarding FinTech influences the acceptance of QRIS. This study applies the UTAUT-3 model to examine how the moderating variables may affect millennials' behavioral intentions and use behaviors toward QRIS. Quantitative methods are employed in this study; 200 samples of millennials in Makassar, South Sulawesi, Indonesia, were collected and analyzed using SEM-PLS. The findings show that performance expectancy, effort expectancy, social influence, hedonic motivation, price value, and technology readiness significantly affect behavioral intention (p < 0.05). Similarly, facilitating conditions, habit, technology readiness, and security and privacy significantly affect use behavior (p < 0.05). However, habit, regulation and policy, and security and privacy did not significantly impact behavioral intention (p ≥ 0.05).
This paper investigates the rate of increase in earth resistance due to corrosion of the earth electrode. When the earth electrode corrodes, the predetermined value cannot be maintained due to the earth resistance increasing. As there has been no study of the relationship between corrosion of the earth electrode and earth resistance, it is difficult to predict the rate of increase in earth resistance due to electrode corrosion. In this paper, we have derived a calculation formula for determining changes in electrode shape that takes both electrolytic corrosion and natural corrosion into account, and this makes it possible to estimate the lifetime of the earth electrode. Furthermore, by using this calculation formula, it is possible to predict the rate of increase in earthing resistance value when electrolytic corrosion and natural corrosion progress simultaneously.
This study looks at how public interventions helps startups succeed in Albania. It focuses on technology growth and independent institution playing role. Drawing on a survey of 80 startups – both subsidized and unsubsidized – we applied k-means clustering, PCA and multiple regression to analyze policy, bureaucratic barriers, innovation and performance functioning. The results show that subsidized startups perform better on average (β = 0.359, p = 0.095), while innovation remains the strongest predictor of success (β = 0.381, p = 0.001). PCA revealed that innovation and success form correlated components, while bureaucratic barriers are a separate dimension. The findings suggest that only a potential impact is underestimated; their effectiveness depends on the innovative capacity of startups and a supportive institutional environment. This study provides the first empirical evidence on public financing of startups in Albania, starting from the policies of an entrepreneurial system in a developing economy.
This study aims to determine the effect of financial literacy, information technology, education level, firm age, owner’s motivation, and firm size on the adoption of SMEs financial accounting standards (SMEs FAS) at the SMEs in the Banyumas region of Indonesia. By using a purposive sampling approach, from around 1,000 SMEs as the population, 130 were selected as a sample and the data was gathered by the distribution of questionnaires. Of 300 questionnaires during May-September 2024, 130 were returned. Then, multiple linear regression analysis was used to analyze the data. The results show that the information technology, education level, motivation, business age, and business size influence positively on the implementation of SMEs FAS, while financial literacy has no effect. With its limitations such as relatively few numbers of samples, this study contributes to developing knowledge in the field of accounting for SMEs, particularly in emerging countries such as Indonesia.
This article proposes the new Nakagami-generated family of distribution (NNak-G) and the new Nakagami-Exponential (NNak-E) distribution for fitting skewed and heavy-tailed data. By integrating the flexibility of the Nakagami distribution with the useful properties of the exponential distribution, the NNak-E distribution provides enhanced adaptability for fitting with right-skewed data structures. The theoretical properties of the NNak-E distribution are developed, including its distribution function, density function, moments, and hazard function. Parameter estimation is constructed using the Maximum Likelihood Estimation (MLE) method, and a comprehensive simulation study is performed to evaluate the performance of the MLE estimates. Furthermore, the NNak-E distribution is applied to real-world datasets, where it demonstrates superior goodness-of-fit compared to existing distributions. The results confirm that NNak-E is a choice distribution for practitioners and researchers dealing with asymmetric and heavy-tailed data distributions in diverse fields such as engineering, medical research, and risk assessment.
The role of SMEs is very large in supporting the economy of a country, including Indonesia. The existence of thousands of SMEs in Bali is expected to improve the welfare of the Balinese people. However, in reality, there are still many craft SMEs in Bali that have not been able to develop their business scale to be larger. This study was conducted to confirm the phenomenon and empirical findings of previous studies on the relationship between mindset and entrepreneurial sustainability. The integration of attitude, subjective norm, and PBC as mediators in the relationship between mindset and SME resilience is the novelty of this study. SMEs based on the creative economy in the Craft subsector in Bali are the population, and the research sample is the owners or leaders of SMEs. Proportional random sampling was chosen as a technique in determining respondents, and SEM-PLS is the analysis method. The findings show that a growth mindset is connected to resilience and the Theory of Planned Behavior (TPB) dimensions. Likewise, an entrepreneurial mindset affects the TPB dimensions, but not resilience. Strengthening mindsets can be a solution to SMEs' resilience in facing dynamics and the capacity to realize business sustainability.
The changing needs of customers for a more interactive, personalized, and smooth shopping experience have changed the field of e-commerce. This has led to the use of immersive technologies like Augmented Reality (AR), Virtual Reality (VR), and Artificial Intelligence (AI), which have a great potential to enrich the consumer experience, increase efficiency, and ultimately contribute to higher sales through the integration of AI. This study aims to examine the influence of AI and immersive technologies on online shopping, integrating contemporary trends, consumer behaviors, and case studies. To achieve this, the study aims to combine theoretical and practical knowledge through the synthesis of 210 sources, later narrowed down to 89 relevant ones. The study provides a unified understanding of a previously fragmented field, creating a valuable source of knowledge for any individual seeking to advance the field of online shopping.
The rapid growth of e-commerce has led to a rise in fraudulent interactions, causing disruptions to digital platforms and the financial ecosystems that support them. The study applied various classification models, including Naïve Bayes, Decision Tree, Random Forest, and K-Nearest Neighbors, and evaluated their performance using key metrics such as accuracy, precision, recall, and F1 score. Model selection was aided by information gain measures for attribute selection to enhance performance. Additionally, model accuracy was improved through hyperparameter tuning and ensemble methods, such as voting across models at the same risk level. Overall, the evaluation showed that classification models, especially well-tuned ones, are effective solutions for e-commerce fraud detection. This research also provides a foundation for developing automated fraud detection systems and supports the ongoing efforts to improve cybersecurity and trust in online commerce.
urpose: In the context of increasingly scarce medical resources, the main purpose of this paper is to propose a SSD-based Variance-based Upsampling and Pyramid Voting (SSD-VUPV) model for improving the efficiency and accuracy of automated detection of lung nodules, as one of the early manifestations of lung cancer. Methods: Firstly, the proposed SSD-VUPV model adapts to the detection of small pulmonary nodules by changing the size of the feature map of the input prediction module. Secondly, the prediction frame is modified to make greater use of the shallow feature layer. In addition, a set of up-Block modules is augmented through the incorporation of asymmetric convolutions, which involves the utilization of a Feature Pyramid Network (FPN) mechanism. Finally, multi-scale and asymmetric convolutions are added to the model to further improve its detection performance. Results: Experimental results, obtained on two public datasets, show that SSD-VUPV can increase the mAP@0.5, F1 score, and sensitivity of the baseline model (SSD) from 63.01% to 87.52%, 64.20% to 90.24%, and 63.25% to 88.74%, on the LUNA16 dataset, and from 77.6% to 83.2%, 76.5% to 83.0%, and 75.8% to 81.9%, on the ChestX-ray14 dataset, respectively. Moreover, SSD-VUPV outperforms state-of-the-art models, based on their results reported in the literature, according to all evaluation metrics used. Conclusion: By cleverly integrating a feature pyramid structure and incorporating the newly designed up_Block modules, the proposed SSD-VUPV model can combine deep semantic features with shallow detail features, thus fully leveraging the rich feature information in the medical images, which allows it to reach top detection accuracy and robustness. Moreover, the inclusion of the newly designed Visual Multi-scale Asymmetric Convolution (VMAC) modules enables the model to adapt to different scale receptive fields, which enables it to capture more varied and detailed features, deepening its understanding of the input data and significantly improving its ability to capture features of various sizes. Consequently, the proposed SSD-VUPV model exhibits improved performance in scenarios involving complex backgrounds and targets.
This paper presents a novel swarm evolutionary metaheuristic optimization algorithm inspired by the cyclic nature of human civilization in its rise from nomadic life to the point where it peaks, stagnates, and then declines back to nomadic life. Nomads do the exploration, while civics do the exploitation. The proposed algorithm also mimics basic attributes and behaviors of human civilizations from hostility that pushes civilizations away from each other to cooperation and trade which helps to explore areas in between and exchange knowledge of best traits (products). Experimental results show highly competitive results when benchmarked with common swarm-based evolutionary algorithms such as Genetic Algorithm, Particle Swarm Optimization, Wolf Pack Algorithm, and Artificial Bee Colony, using common test functions like Rastrigin, Schwefel, Rosenbrock, Griewank, and Ackley.
Complex multicomponent materials currently form the basis of nano- and microelectronics, space technology, information storage and transmission systems, intelligent systems, and other fields of science and engineering. However, describing the microstructure of materials, which is formed as a result of multiparticle interactions, as well as reliably predicting their physical properties, remain open questions. This is explained by the fact that existing theoretical models do not adequately describe multiparticle interactions. This article presents a general overview of models developed by the authors to describe the simultaneous processes of swarming and multiparticle aggregation in both batch and through-flowing devices. To address this problem, both a new stochastic lattice model (SLM) and software for its numerical implementation have been carried out. The article presents the results of testing the new methods in the computer experiment. The model and software can be useful in creating methods for calculating various technological equipment and processes.
This paper presents a literature review of key business needs that remain difficult to address through traditional digital systems but can be effectively supported by blockchain technology. A total of 356 research papers were analyzed using a BERT-based clustering approach, leading to the identification of five major need categories: transparency and trust, data integrity and provenance, fraud prevention in identity management, fractional ownership, and secure cross-border payments. The implementation strategies associated with these needs are linked to practical use cases across sectors such as agriculture, supply chain, fintech, real estate, and e-healthcare. The findings highlight a growing shift from Layer-1 to more scalable Layer-2 solutions, driven by lower costs and increased development flexibility. In addition, performance benchmarks and regulatory developments in the EU, US, Japan, and Singapore are examined. The study concludes by proposing an iterative adoption framework in which business needs, technological choices, and regulatory considerations continuously inform one another.
This study investigates how to apply a multi-agent system (MAS)-based process on robotic platforms to mine detection, especially improving collaboration/communication between agents. Agents can communicate and collaborate by sharing sensor data relevant to the task, sharing requests for tasks to be acted on by agents, applying standard protocols, wireless communications, message passing, etc. Using a publish-subscribe architecture will enable the agents not to force all agents to receive all information and use decentralized decision-making to promote adaptable behavior. In addition, through adaptive communications protocols, a better means of data sharing, adaptations of behavior, and higher-order activities will be simpler. Finally, a shared state of relevant communication will allow collaborative robotic agents in an MAS to have higher accuracy and efficiency to clear minefields while improving safety and effectiveness.
This article proposes a new distribution called the mixture Nakagami-Rayleigh (mNR) distribution, which is obtained by combining Nakagami and Rayleigh distributions. The maximum likelihood estimator (MLE) is the primary approach we propose to estimate the parameters for the mN-R distribution. This paper also presents related functions of the mNR distribution, such as the survival function, the probability density function (PDF), the cumulative distribution function (CDF), and the hazard function, along with the important mathematical properties of the mNR distribution. Additionally, plots of the important functions of the mNR distribution are illustrated as line plots and contour plots. Furthermore, we present a simulation study to demonstrate the flexibility of the mNR distribution. Finally, the article concludes by summarizing the key outcomes and providing recommendations for future work.
Smart parking and traffic management are some of the major problems faced during the development of a smart city. There exist numerous challenges when trying to develop systems to solve the above-mentioned problem since Vehicular Ad-Hoc Networks (VANETs) produce a large volume of spatiotemporal data (vehicle speed, direction, and GPS position) in real-time. Nevertheless, due to their complexity, it is difficult to detect the existence of any abnormal events rapidly and reliably. In order to tackle such challenges, we suggest developing a hybrid deep neural network model, integrating Long-Short Term Memory (LSTM) network capable of capturing long-term dependencies and Squeeze-and-Excitation block (attention mechanism) highlighting important features. The proposed model is tested on AV-GPS-Dataset, containing GPS tracks of real vehicles as well as attack instances. Experiments show excellent results (accuracy equals to 99.81%, precision to 99.81%, recall to 99.73%, and F1 score to 99.74%), which outperform traditional methods significantly. Our results confirm that spatiotemporal data plays a vital role in solving problems stated.
