Multi-stage Approach Research Articles

Cocoa Farmers’ Perceptions of Drought and Adaptive Strategies in the Ghana–Togo Transboundary Cocoa Belt

This study investigated the perception of drought by cocoa farmers and explored the effectiveness of adaptive strategies (ASs) used in smallholding farms in the transboundary region between Ghana and Togo. Drought significantly threatens cocoa production in this region, affecting farmers’ livelihoods and cocoa supply chains. This study used a multistage sampling approach, which involved surveys with questionnaires administered to 330 cocoa farmers throughout the study area, along with on-site observations. Statistical analysis included binary logistic and Poisson regression models to explore the relationship between farmer socioeconomic characteristics and adaptation practices. The findings revealed that cocoa farmers in the region have a nuanced understanding of drought, attributed to changing climatic patterns and unsustainable land management practices such as deforestation. To mitigate its impacts, farmers employ a variety of ASs, including investment in farm management, soil management, and intercropping with crop diversification. Furthermore, socioeconomic factors, including age, formal education, household size, land tenure right, adaptation cost assessment, and an underestimation of self-efficacy, were shown to affect the choice in the AS. Among the ASs adopted, only farm management practices (weeding, pruning, fertilizer application, etc.) significantly improved yield. This study contributes to understanding drought as a critical issue for cocoa farmers and the adaptation practices used by smallholder cocoa farmers. Given that among the strategies adopted, only farm management practices, also known as good agricultural practices (GAPs), significantly improves yield, this study recommends well-designed and innovative packages of sustainable farm management based on farm and owner characteristics. These include irrigation schemes, timely soil fertilizer monitoring and supply, and the provision of drought-resistant varieties along with technical itineraries. Additional interventions require drought emergency responses, with other factors such as education and financial support mechanisms expected to improve farmers’ timely decision-making to adapt and improve cocoa production resilience to drought episodes in international transboundary regions with complex governance structures.

Strain selection and adaptation of a fungal-yeast-microalgae consortium for sustainable bioethanol production and wastewater treatment from livestock wastewater

This study explores the potential of strain selection and adaptation for developing a fungi-yeast-microalgae consortium capable of integrated bioethanol production and livestock wastewater treatment. We employed a multi-stage approach involving isolation and strain selection/adaptation of these consortiums. The study started with screening some isolated fungi to grow on the cellulosic biomass of the livestock wastewater (saccharification) followed by a fermentation process using yeast for bioethanol production. The results revealed that Penicillium chrysogenum (Cla) and Saccharomyces cerevisiae (Sc) produced a remarkable 99.32 ppm of bioethanol and a concentration of glucose measuring 0.56 mg ml− 1. Following the impact of fungi and yeast, we diluted the livestock wastewater using distilled water and subsequently inoculated Nile River microalgae into the wastewater. The findings demonstrated that Chlorella vulgaris emerged as the dominant species in the microalgal community. Particularly, the growth rate reached its peak at a 5% organic load (0.105385), indicating that this concentration provided the most favorable conditions for the flourishing of microalgae. The results demonstrated the effectiveness of the microalgal treatment in removing the remaining nutrients and organic load, achieving a 92.5% reduction in ammonia, a 94.1% reduction in nitrate, and complete removal of phosphate (100%). The algal treatment also showed remarkable reductions in COD (96.5%) and BOD (96.1%). These findings underscore the potential of fungi, yeast, and Nile River microalgae in the growth and impact on livestock wastewater, with the additional benefit of bioethanol production.Graphical abstract

Correlation between hyperlipidemia and serum vitamin D levels in an adult Chinese cohort.

Vitamin D deficiency has emerged as a significant concern in public health due to its potential association with various metabolic disorders. This study aimed to investigate the relationship between serum vitamin D levels and the susceptibility to hyperlipidemia among adults. Using a multi-stage sampling approach, we recruited a cohort of 2072 eligible individuals aged over 18 years. Serum 25-hydroxyvitamin D [25(OH)D] levels were measured alongside glucolipid metabolic markers, and comprehensive demographic and physical data were collected. The cohort exhibited a hyperlipidemia prevalence of 42.18%, with 19.88% demonstrating vitamin D deficiency. Notably, 23.68% of individuals with vitamin D deficiency also presented hyperlipidemia. Statistical analysis revealed a significantly higher prevalence of hyperlipidemia among those with vitamin D deficiency compared to those with sufficient levels (23.68% vs. 17.11%, P < 0.05). After adjusting for various factors including age, geographical region, exercise status, BMI, fasting glucose level, and blood pressure, lower serum 25(OH)D concentrations were found to significantly increase the risk of hyperlipidemia (Odds Ratio [OR] = 1.41; 95% CI: 1.057, 1.885; P < 0.05). Further stratification of the hyperlipidemic cohort revealed that vitamin D deficiency was associated with 1.459- and 1.578-times higher risks for total cholesterol and triglyceride abnormalities, respectively, compared to those with sufficient vitamin D levels. Moreover, each 10 ng/mL decrease in serum vitamin D level corresponded to an increased risk of total cholesterol (OR = 0.82; 95% CI: 0.728, 0.974; P < 0.05) and triglyceride abnormalities (OR = 0.79; 95% CI: 0.678, 0.927; P < 0.05). However, there were no significant differences observed between vitamin D-sufficient and-deficient groups regarding Low-Density Lipoprotein (LDL) and High-Density Lipoprotein (HDL) abnormalities. These findings underscore the potential role of serum vitamin D deficiency as an independent risk factor contributing to the elevated prevalence of hyperlipidemia in the adult population.

Hard jet substructure in a multistage approach

We present predictions and postdictions for a wide variety of hard jet-substructure observables using a multistage model within the framework. The details of the multistage model and the various parameter choices are described in []. A novel feature of this model is the presence of two stages of jet modification: a high-virtuality phase [modeled using the modular all twist transverse-scattering elastic-drag and radiation model ()], where modified coherence effects diminish medium-induced radiation, and a lower virtuality phase [modeled using the linear Boltzmann transport model ()], where parton splits are fully resolved by the medium as they endure multiple scattering induced energy loss. Energy-loss calculations are carried out on event-by-event viscous fluid dynamic backgrounds constrained by experimental data. The uniform and consistent descriptions of multiple experimental observables demonstrate the essential role of modified coherence effects and the multistage modeling of jet evolution. Using the best choice of parameters from [], and with no further tuning, we present calculations for the medium modified jet fragmentation function, the groomed jet momentum fraction zg and angular separation rg distributions, as well as the nuclear modification factor of groomed jets. These calculations provide accurate descriptions of published data from experiments at the Large Hadron Collider. Furthermore, we provide predictions from the multistage model for future measurements at the BNL Relativistic Heavy Ion Collider. Published by the American Physical Society 2024

A Comprehensive Survey on Deep Clustering: Taxonomy, Challenges, and Future Directions

Clustering is a fundamental machine learning task which aims at assigning instances into groups so that similar samples belong to the same cluster while dissimilar samples belong to different clusters. Shallow clustering methods usually assume that data are collected and expressed as feature vectors within which clustering is performed. However, clustering high-dimensional data, such as images, texts, videos, and graphs, poses significant challenges for clustering tasks, such as indiscriminate representation and intricate relationships among instances. Over the past decades, deep learning has achieved remarkable success in effective representation learning and modeling complex relationships. Motivated by these advancements, Deep Clustering seeks to improve clustering outcomes through deep learning techniques, garnering considerable interest from both academia and industry. Despite many contributions to this vibrant area of research, the lack of systematic analysis and a comprehensive taxonomy has hindered progress in this field. In this survey, we first explore how deep learning can be integrated into deep clustering and identify two fundamental components: the representation learning module and the clustering module. Then we summarize and analyze the representative design of these two modules. Furthermore, we introduce a novel taxonomy of deep clustering based on how these two modules interact, specifically through multistage, generative, iterative, and simultaneous approaches. In addition, we present well-known benchmark datasets, evaluation metrics, and open-source tools to clearly demonstrate different experimental approaches. Finally, we examine the practical applications of deep clustering and propose challenging areas for future research.

Bearing remaining life prediction method based on ARAD -ELN and multi-stage wiener process

Abstract Stochastic process-based models are extensively utilized in health assessments and Remaining Useful Life (RUL) predictions of bearings. Nevertheless, bearings in actual operation undergo multiple degradation stages, each characterized by its unique trend of degradation. The application of a singular stochastic process for RUL prediction falls short of achieving optimal performance. Consequently, this paper introduces a multi-stage Wiener process-based approach for the prediction of bearings' RUL. Initially, to tackle the challenge of imbalanced sample sizes across different degradation stages of bearings, an ensemble learning-based neural network (ELN) enhanced by ARIMA Residual Anomaly Detection (ARAD) for the identification of bearing degradation stages is proposed. Subsequently, taking into account the temporal, unit-to-unit, and nonlinear variabilities of the degradation process at each stage, a Wiener process-based multi-stage degradation model for bearings is developed. A method for parameter estimation and updating, utilizing Kalman filtering and Maximum Likelihood Estimation (K-M) is introduced. Finally, the proposed model undergoes validation using both simulated data and the XJTU-SY bearing dataset. Experimental results of three RUL predictions show that the proposed method leads the benchmark model with RMSE values of 3.61, 2.92 and 7.24 respectively, affirming that the proposed model can precisely classify equipment degradation stages and predict RUL with high accuracy and stability.&amp;#xD;

Raising Awareness of the Impact of Stunting on Toddler Development Through Parental Education

In 2023, the prevalence of stunting in Indonesia remains at 21.5%, showing only a marginal decrease of 0.1% from the previous year and still exceeding the WHO standard of 20%. Stunting adversely affects children, leading to impaired physical and cognitive development, increased risk of developmental disorders, and a higher likelihood of metabolic diseases in adulthood.This study utilized a multi-stage approach: data and area analysis, educational and developmental screening, and evaluation. Participants included 44 parents with their toddlers, 5 cadres, and 3 village officials. Knowledge among participants improved significantly, with 76% demonstrating enhanced understanding. Statistical analysis revealed a significant increase in knowledge post-intervention (p=0.000). Developmental assessments showed that 11% of toddlers were classified as "doubtful" and 23% as having "deviant" results. Parents of these children received education on age-appropriate stimulation, and cases were reported to local health workers. The goal is for the acquired knowledge to be widely applied, with ongoing routine screening for early detection and timely management of developmental issues.

Using 3D Hand Pose Data in Recognizing Human–Object Interaction and User Identification for Extended Reality Systems

Object detection and action/gesture recognition have become imperative in security and surveillance fields, finding extensive applications in everyday life. Advancement in such technologies will help in furthering cybersecurity and extended reality systems through the accurate identification of users and their interactions, which plays a pivotal role in the security management of an entity and providing an immersive experience. Essentially, it enables the identification of human–object interaction to track actions and behaviors along with user identification. Yet, it is performed by traditional camera-based methods with high difficulties and challenges since occlusion, different camera viewpoints, and background noise lead to significant appearance variation. Deep learning techniques also demand large and labeled datasets and a large amount of computational power. In this paper, a novel approach to the recognition of human–object interactions and the identification of interacting users is proposed, based on three-dimensional hand pose data from an egocentric camera view. A multistage approach that integrates object detection with interaction recognition and user identification using the data from hand joints and vertices is proposed. Our approach uses a statistical attribute-based model for feature extraction and representation. The proposed technique is tested on the HOI4D dataset using the XGBoost classifier, achieving an average F1-score of 81% for human–object interaction and an average F1-score of 80% for user identification, hence proving to be effective. This technique is mostly targeted for extended reality systems, as proper interaction recognition and users identification are the keys to keeping systems secure and personalized. Its relevance extends into cybersecurity, augmented reality, virtual reality, and human–robot interactions, offering a potent solution for security enhancement along with enhancing interactivity in such systems.

Traffic Flow Prediction in 5G-Enabled Intelligent Transportation Systems Using Parameter Optimization and Adaptive Model Selection.

This study proposes a novel hybrid method, FVMD-WOA-GA, for enhancing traffic flow prediction in 5G-enabled intelligent transportation systems. The method integrates fast variational mode decomposition (FVMD) with optimization techniques, namely, the whale optimization algorithm (WOA) and genetic algorithm (GA), to improve the accuracy of overall traffic flow based on models tailored for each decomposed sub-sequence. The selected predictive models-long short-term memory (LSTM), bidirectional LSTM (BiLSTM), gated recurrent unit (GRU), and bidirectional GRU (BiGRU)-were considered to capture diverse temporal dependencies in traffic data. This research explored a multi-stage approach, where the decomposition, optimization, and selection of models are performed systematically to improve prediction performance. Experimental validation on two real-world traffic datasets further underscores the method's efficacy, achieving root mean squared errors (RMSEs) of 152.43 and 7.91 on the respective datasets, which marks improvements of 3.44% and 12.87% compared to the existing methods. These results highlight the ability of the FVMD-WOA-GA approach to improve prediction accuracy significantly, reduce inference time, enhance system adaptability, and contribute to more efficient traffic management.

Initial monitoring of a six-story lightweight timber frame building under different environmental conditions

AbstractThis study presents a comprehensive investigation of the dynamic behavior of a six-story lightweight timber frame building through periodic and continuous ambient vibration tests. Seasonal changes in temperature and relative humidity can influence the dynamic response of timber buildings, by affecting stiffness, strength, and connection properties due to changes in moisture content. Systematic tests were performed from October 2022 until May 2024 to identify natural frequencies, damping ratios, and mode shapes of the building using five battery-driven data acquisition units equipped with 15 uni-axial accelerometers. Two different only-output frequency and time domain Operational Modal Analysis (OMA) methods were used to evaluate the dynamic properties of the building. Additionally, the building has been continuously monitored with temperature and humidity sensors since its construction. Results obtained from the periodic measurements highlighted the need for a permanent system to capture the transient changes in the modal parameters. A complementary permanent system to record the accelerations at the roof level was installed in May 2024. The modal parameters from in situ measurements were compared with those obtained from the Finite Element (FE) model of the structure. The FE model, calibrated through a multi-stage approach, incorporating non-structural elements, and varying imposed loads, showed good agreement with experimental data. Comparative analysis of the results obtained under different temperature and humidity conditions showed the effect of the environmental conditions on the building dynamics properties, for both periodic and permanent measurements. This study highlights the importance of continuous monitoring and detailed FE modeling in understanding the dynamic properties and long-term structural behavior under varying conditions. This information can be used to ensure that the building meets safety and performance requirements and to identify potential issues that may need to be addressed during the design and maintenance phases.

Public awareness of gastric cancer risk factors and screening behaviours in Shijiazhuang, China: A community-based survey.

Reducing exposure to risk factors and screening represent 2 major approaches to gastric cancer (GC) prevention, but public knowledge GC risk factors and screening behaviour remain unknown. We aimed to investigate public awareness of GC risk factors, adherence to screening, and barriers hindering screening practices in China. This community-based household survey was conducted within Shijiazhuang, China, and 1490 residents were recruited through a multistage stratified cluster random sampling approach. A self-administered questionnaire was completed which consisted of three sections: demographics, awareness of GC risk factors, and personal screening behaviours. Factors associated with knowledge of risk factors and screening behaviours were evaluated using binary logistic regression analysis. The mean risk factor awareness score of 12 (7, 15) revealed insufficient knowledge in 51.1% of participants. Dietary lifestyle factors were better understood than physical activity and weight-related factors. Marital status (OR 1.967; 95% CI 1.415 to 2.734), higher income (OR 1.197; 95% CI 1.010 to 1.418), and a history of upper gastrointestinal problems (OR 0.048; 95% CI 1.002 to 1.311) were associated with higher awareness. Merely 21.5% underwent GC screening, with higher rates linked to older age (OR 1.642; 95% CI 1.418 to 1.902), higher education (OR 1.398; 95% CI 1.176 to 1.662), a history of upper gastrointestinal problems (OR 3.842; 95% CI 2.833 to 5.209), and moderate (OR 2.077; 95% CI 1.352 to 3.191) and high (OR 2.529; 95% CI 1.311 to 4.878) perceived GC risk. Notably, participants commonly refused gastroscopy due to the absence of symptoms or signs. In Shijiazhuang, more than half of participants demonstrated inadequate knowledge of GC risk factors, and screening participation rates were remarkably low. This emphasizes the need for targeted interventions to enhance GC awareness and significantly improve screening rates.

Stance Detection in the Context of Fake News—A New Approach

Online social networks (OSNs) are inundated with an enormous daily influx of news shared by users worldwide. Information can originate from any OSN user and quickly spread, making the task of fact-checking news both time-consuming and resource-intensive. To address this challenge, researchers are exploring machine learning techniques to automate fake news detection. This paper specifically focuses on detecting the stance of content producers—whether they support or oppose the subject of the content. Our study aims to develop and evaluate advanced text-mining models that leverage pre-trained language models enhanced with meta features derived from headlines and article bodies. We sought to determine whether incorporating the cosine distance feature could improve model prediction accuracy. After analyzing and assessing several previous competition entries, we identified three key tasks for achieving high accuracy: (1) a multi-stage approach that integrates classical and neural network classifiers, (2) the extraction of additional text-based meta features from headline and article body columns, and (3) the utilization of recent pre-trained embeddings and transformer models.

A multi-stage machine learning approach for stock price prediction: Engineered and derivative indices

In this paper, a machine learning approach is proposed to predict the next day's stock prices. The methodology involves comprehensive data collection and feature generation, followed by predictions utilizing Multi-Layer Perceptron (MLP) networks. We selected 5,283 records of daily historical data, including open prices, close prices, highest prices, lowest prices, and trading volumes from four well-known stocks in the FTSE 100 index. A novel set of engineered and derivative indices is extracted from the original time series to enhance prediction accuracy. Two Multi-Layer Perceptron (MLP) are proposed to predict the next day's stock prices using the engineered discrete and continuous indices. The case study uses the daily historical time series of stock prices between January 1, 2000, and December 31, 2020. The proposed machine learning approach presents suitable applicability and accuracy, respectively.

Optimising building performance for a resilient Future: A Multi-Objective approach to Net Zero energy strategies

The escalating impacts of climate change on urban environments underscore the imperative for energy-efficient buildings. As the building sector constitutes approximately 40% of global energy consumption, the implementation of NZE designs is critical for emission reduction. This study utilised Building Performance Optimisation methodologies alongside a multi-stage optimisation framework to evaluate energy consumption, Life cycle cost and carbon emissions under various climate scenarios (RCP 2.6, 4.5, and 8.5) in three Algerian cities—Constantine (semi-arid), Algiers (Mediterranean), and Ghardaïa (arid).The results revealed that the integration of passive, active, and renewable energy strategies significantly curtailed energy consumption and CO2 emissions across different climatic contexts. Specifically, passive measures achieved reductions in energy use ranging from 39% in Ghardaïa to 15% in Constantine. Active systems, including high-efficiency HVAC technologies, resulted in energy savings of 59% in Ghardaïa, 52% in Algiers, and 47% in Constantine. Furthermore, renewable energy sources, with a focus on solar photovoltaics, contributed additional reductions: 44% in Algiers, 26% in Ghardaïa, and 20% in Constantine. The multi-stage optimisation approach also enhanced computational efficiency, significantly reducing simulation time and facilitating the rapid identification of optimal strategies.These findings highlight the efficacy of integrating a range of energy strategies to achieve NZE. Nevertheless, while these systems demonstrate robust performance under current conditions, their effectiveness diminishes under future climate scenarios, underscoring the necessity for adaptive and resilient design approaches. Ongoing optimisation and innovation will be essential to maintaining the efficacy of NZE strategies in the face of extreme future climate conditions.

A comparative study of evolutionary algorithms and particle swarm optimization approaches for constrained multi-objective optimization problems

Many real-world optimization problems contain multiple conflicting objectives as well as additional problem constraints. These problems are referred to as constrained multi-objective optimization problems (CMOPs). Many meta-heuristics for solving CMOPs, called constrained multi-objective meta-heuristics (CMOMHs) have been introduced in the literature, including those using particle swarm optimization (PSO)(Kennedy and Eberhart, 1995), genetic algorithms (GAs)(Man et al., 1996), and differential evolution (DE)(Storn and Price, 1997). CMOMHs can be grouped into four different classes: classic CMOMHs, co-evolutionary approaches, multi-stage approaches, and multi-tasking approaches. An extensive comparative study of twenty different CMOMHs on a wide variety of test problems, including real-world CMOPs in the fields of science and engineering, is conducted. A multi-swarm PSO approach called constrained multi-guide particle swarm optimization (ConMGPSO) is introduced and compared to the best-performing previous approaches according to the comparative study. The performance of each algorithm was found to be problem dependent, however the best overall approaches were ConMGPSO, paired-offspring constrained evolutionary algorithm (POCEA)(He et al., 2021), adaptive non-dominated sorting genetic algorithm III (A-NSGA-III)(Jain and Deb, 2014), and constrained multi-objective framework using Q-learning and evolutionary multi-tasking (CMOQLMT)(Ming and Gong, 2023). ConMGPSO and POCEA had the best performance on the CF benchmark set, which contains examples of bi-objective and tri-objective CMOPs with disconnected CPOFs. The CMOQLMT approach had the best performance on the DAS-CMOP benchmark set, which contain additional difficulty in terms of feasibility-, convergence-, and diversity-hardness. For the selected real-world CMOPs, A-NSGA-III had the best performance overall. ConMGPSO was shown to have the best performance on the process, design, and synthesis problems, and had competitive performance for the power system optimization problems.

Risk factors underlining reproductive performance in smallholder beef cattle herds of South Africa

Improving the historically low reproductive performance of beef cattle in smallholder systems is essential for its productivity and profitability. Therefore, identifying and addressing risk factors associated with low performance in this system present an opportunity for improvement. The study aimed to evaluate the effect of animal and management risk factors on pregnancy rate (PR), fetal and calf loss (FC), calving interval (CI) and days open (DO) in smallholder beef cattle farms. A multi-stage selection approach was conducted in five provinces of South Africa. Cow records (3694) collected from 40 smallholder herds over two years (2018–2019) were analysed. Data on animal and herd management factors including body condition score (BCS), cow age class, breed type, lactation status, culling old/non-productive cows, record-keeping, and breeding and calving seasons were recorded. The GLIMMIX procedure was computed to determine risk factors associated with performance indicators (PR, FC, CI and DO). Risk factors highly associated with performance were breed type, BCS, cow-age class and breeding/calving season (P ≤ 0.05). Indicators FC, DO and CI increased with decreased BCS, autumn calving seasons, and in first calvers and aged cows. Whereas, PR increased with increasing odds of BCS and breeding seasons between December-March, November-February and January-March. Optimizing reproductive performance in smallholder beef cattle herds may involve managing cow body condition, breeding with adapted genotypes and synchronizing breeding with favourable seasons. Therefore, record keeping and regular monitoring of herd nutrition, climate and breed performance may be significant in improving reproductive performance in smallholder herds.

A multi-stage approach for desired part grasping under complex backgrounds in human-robot collaborative assembly

Human-robot collaborative assembly can leverage the unique capabilities of humans and robots to provide a flexible and efficient way for complex tasks. In this context, robots are expected to be endowed with the perception ability to collaborate with humans. For flexible processes, robotic grasping for a desired assembly part from variable multi-parts is essential but remains challenging, especially in situations of complex backgrounds including disordered placement, occlusion, similarity, and large-scale differences in size. To improve the perception and decision-making abilities of robots in collaborative assembly, firstly, a multi-stage approach integrating 2D part recognition and 6D pose estimation with perspective-focusing is proposed for desired part grasping under complex backgrounds. Secondly, to efficiently and accurately recognize a desired part from variable multi-parts under complex backgrounds, based on RGB data, an enhanced detector is adopted, which lays the foundation for the subsequent pose estimation. Thirdly, based on point cloud data, a pose estimation method based on perspective-focusing is proposed to enhance the success rate and efficiency of pose acquisition. Additionally, a modular autonomous robotic grasping system is designed. Finally, taking decelerator assembly as a case, the proposed approaches are comprehensively validated and compared on a real collaborative robot platform. Under complex backgrounds with occlusion, our approach shows stable performance with an average grasping success rate of 90.0 % and an average cycle time of 6.4175 s. Our work is expected to improve robots’ perception and decision-making capabilities while further reducing the cycle time of collaborative assembly tasks.

Classification of Shoulder Implant Manufacturer Using Pre-Trained DenseNet201 Combined With Capsule Network.

This study aims to accelerate revision surgery and treatment using X-ray imaging and deep learning to identify shoulder implant manufacturers in advance. A feature engineering approach based on principal component analysis and a k-means algorithm was used to cluster shoulder implant data. In addition, a pre-trained DenseNet201 combined with a capsule network (DenseNet201-Caps) shoulder implant classification model was proposed. DenseNet201-Caps was the most effective classification model on the clustered dataset with an accuracy of 94.25% and an F1 score of 96.30%. Notably, clustering the dataset in advance improved the accuracy and the Caps implementations successfully enhanced the performance of all convolutional neural network models. The analysed results indicate that DenseNet201-Caps struggled to distinguish between the Cofield and Depuy manufacturers. Hence, a multistage classification approach was developed with an improved accuracy of 96.55% achieved. The DenseNet201-Caps method enables the accurate identification of shoulder implant manufacturers.

Multi-stage control design for oscillating water column-based ocean wave energy conversion system

Despite the enormous global potential of ocean wave energy, it has yet to achieve a level of maturity and economic competitiveness that would result in a substantial impact. Challenges include direct integration into weak or isolated microgrids, a high proportion of uncertain marine environments, nonlinear dynamics, oscillating water column (OWC) device limitations, slower response times, unplanned power outages, power fluctuations, high capital and operational costs in ocean wave energy conversion (OWEC) systems. To this end, a new independent multi-stage design approach is proposed for the performance enhancement of an OWC-based OWEC system. Firstly, an airflow and rotational speed optimal control stage enhances power capture in the Wells turbine-based OWC plant. Secondly, compared to conventional control, the proposed permanent magnet synchronous generator control incorporates an adaptive nonlinear back-stepping control algorithm based on Lyapunov stability theory. Thirdly, introducing reconfigurable control into the conventional six-leg power converter ensures the uninterrupted operation of an OWEC system. Lastly, a model-predictive control-based energy management system is integrated with a bidirectional DC-DC converter that delivers steady power from the grid-connected OWC OWEC system. Hence, MATLAB simulations ensure the overall performance enhancement and feasibility of the OWEC system application and verify that the proposed multi-stage solution is efficient, robust, and reliable.

Exploring the impact of integrated design on employee learning engagement in the ubiquitous learning context: A deep learning-based hybrid multistage approach

Learning engagement has received the attention of academics and practitioners; however, studies on employee learning engagement are limited. Based on an integrated hardware-software-value-design––– perspective and domain-specific innovativeness theory, we developed and tested a theoretical framework using a novel and hybrid multistage approach combining a partial least squares (PLS) structural equation model (SEM) and artificial neural networks from deep learning. We used multigroup analysis (PLS-MGA-ANN), which examines key integrated design elements and domain-specific innovativeness drivers of employee learning engagement in toward ubiquitous learning context. According to a sample of learners’ responses, the linear PLS-SEM results demonstrated that (a) integrating design elements, including perceived compatibility, familiarity, value, and user interface design, had a direct impact on domain-specific innovativeness; (b) domain-specific innovativeness had a direct impact on employee learning engagement and played a mediating role in the relationship between integrated design elements and employee learning engagement; and (c) copresence moderated the relationships between domain-specific innovativeness and employee learning engagement. Furthermore, through the evaluation of nonlinear models of the neural network, perceived compatibility and value revealed nonlinear average importance. Practical and theoretical implications are discussed.