The abstract in an era where healthcare demands precision, and personalized solutions, “OpenHealth” emerges as a ground breaking accessibility initiative at the intersection of technology and medicine. This comprehensive project focuses on Multi-Disease Detection, employing a diverse set of algorithms, encompassing deep learning, standard machine learning, transfer learning, and hybrid models such as VGG-19, ResNet50, Random Forest (RF), and Gradient Boosting. Diseases across specific organs, such as the brain, kidney, heart, liver, and lungs, are accurately predicted, and model performance is rigorously assessed through metrics like accuracy, recall. Multi-disease detection enables simultaneous identification of multiple conditions, reducing diagnostic time and supporting early detection of comorbidities in clinical decision-making scenarios. Adding a layer, “OpenHealth” integrates with large language models from the Open-source libraries like Hugging Face, providing personalized information based on individual health profiles. Likewise, the design extends its impact by incorporating an AI dietitian and food recommender, acclimatizing salutary recommendations to individual health conditions. scrupulous association is assured through devoted directory structures, fostering a modular and justifiable frame. Leveraging Machine Learning Operations (MLOps) like Dockers, DVC, Evidently, and MLflow enhances the overall efficiency and reliability of healthcare systems.
This study evaluates the effect of HRM practices on the job satisfaction levels of employees working for PSUs located in Telangana, India. A structured questionnaire survey method has been used to gather data on the human resource policies of organizations in relation to employee involvement and job satisfaction levels, among other variables. The data obtained was then subjected to statistical analysis, using correlations, structural models and trends among others. Structural Equation Modelling was further applied in testing the relationships between the HRM practices, employee engagement and job satisfaction. In addition to that, Confirmatory Factor Analysis was also conducted to validate the measurement model and assess its reliability and validity. According to the findings of this study, there exists a significant relationship between HRM practices and job satisfaction, with employee engagement having a powerful mediating role. The structural equation model showed good model fit, with acceptable indices (RMSEA = 0.04, CFI = 0.90, and TLI = 0.88). The use of these fit indices in the study shows that the analytical approach used was valid and reliable. Therefore, the findings indicate that the application of proper HRM practices such as development plans, fair treatment and good performance appraisals increases job satisfaction among.
Predicting student performance by demographic data and assignment grades is decisive for executing effective interceptions that enhance instructional outcomes. This research deals with the query of accurately forecasting student performance to ameliorate learning results and ensure timely support. On applying DL techniques, the SeqGatNet model is developed by combining Graph Neural Networks (GNN) with sequential frameworks to capture both temporal and spatial dependencies that affect student performance over time. This research proposes a hybrid DL model that amalgamates graph-based data processing with sequence modelling, offering a more accurate and reliable system for performance prediction. The framework is evaluated by utilising the student performance dataset and correlating it with other methods, including KNN, XGBoost, and Linear Regression. The framework is evaluated on the UCI Student Performance and OULAD datasets, which include diverse academic and behavioral attributes. These datasets enable robust assessment across varying educational contexts. The proposed SeqGatNet model achieves an accuracy of 99.3% with improved predictive performance over baseline methods. The results manifest that SeqGatNet exceeds these traditional frameworks in accuracy (99.3%), and precision (99.5), with an RMSE of 1.23, emphasizing its enhanced ability to predict student performance. To provide timely interventions and maximize student outcomes, the proposed method assists by providing a scalable solution for real-time educational decision-making. Machine learning transforms business management by enabling predictive analytics, automation, and data-driven decision-making, improving efficiency, accuracy, and organizational performance outcomes.
Background: Agriculture is challenged to satisfy rising food needs in the face of an expanding world population, in addition to environmental and genetic limitations that influence crop yield and nutritional quality. Machine Learning (ML) and Deep Learning (DL) are solving these problems as they enable the accurate establishment of predictions based on large datasets in the field of agriculture. The current models do not consider complex interactions between the treatment, genetic, and environmental variables. Also present of models are prone to fail in generalizing in diverse agricultural environments and cannot be interpreted by the end-users such as farmers. Proposed Method: The study provides a BioAgriNet model, a Honey Badger-Optimized SAINT-BiLSTM that integrates the genotype, environment, and treatments to predict crop yield and nutrient quality. BioAgriNet is an opportunity that exploits the hybrid architecture in the form of Self-Attention and Inters ample Attention Transformer (SAINT) combined with BiLSTM with Honey Badger Optimization (HBO) to select the features, and thus is a stable, interpretable, and efficient model that will be applied in agricultural prediction. Unlike the traditional methods, it not only provides the prediction of the yield and quality of the nutrients but also it provides the contribution to the environmental and genetic factors. Findings: The R 2 of the model was 0.985 which explained 98.5 percent of the variance in yield. RMSE was 0.005, and MAE was 0.03, showing high accuracy in prediction. MAPE was 2.5%, reflecting minimal percentage error. BioAgriNet performed better than comparative baseline models like CNN-LSTM, which had lower R 2 values and higher error rates, reflecting its better predictive ability. Edge Over This Domain: The new model enhances the accuracy of predictions, explainability, and real-time usage, and is an important asset in sustainable agriculture, improving food security, and maximizing market value.
Employee retention and engagement are key drivers of organizational performance. However, traditional approaches frequently miss the complexities of contemporary work patterns. This paper outlines a framework that uses customized e-commerce solutions powered by predictive analytics to improve employee retention and engagement. The framework employs a hybrid model combining Convolutional Neural Networks (CNNs) and Autoencoders. It makes predictions from comprehensive datasets that include employee behavior, performance, and engagement levels. Through a hybrid two-stage feature extraction pipeline combining PCA-based linear dimensionality reduction and CNN-Autoencoder-based nonlinear feature extraction, the model identifies engagement levels and predicts disengagement. The model performance was assessed using several key metrics. The results are: Accuracy [Formula: see text] 99.48%, Precision [Formula: see text] 99.34%, Recall [Formula: see text] 99.63%, F1-Score [Formula: see text] 99.49%, MAE [Formula: see text] 0.0052, RMSE [Formula: see text] 0.0719, and [Formula: see text]. These findings highlight the model’s strong predictive capability for Human Resources (HR) decision-making. It enables targeted, data-driven interventions that improve employee satisfaction, reduce turnover, and boost productivity. The model shifts HR practices from reactive to proactive. This helps organizations retain top talent and maintain a high-performing, motivated workforce in a competitive labor market.
Fabric fault detection in textile manufacturing requires accurate and efficient methods to ensure reliable quality control under varying production conditions. Image and signal processing techniques are utilized to enhance defect-related features through noise reduction, contrast adjustment, and improved data representation for analysis. An integrated approach combining feature enhancement and K-medoids clustering is employed to classify defective and non-defective fabric regions, providing robustness against noise, variability, and outliers commonly present in industrial environments. The clustering process relies on representative medoid patterns, enabling consistent separation of fault characteristics from normal fabric structures. A point-to-multipoint communication mechanism is incorporated to support efficient transmission of inspection data from acquisition units to monitoring systems, facilitating real-time fault detection and analysis. Experimental evaluation conducted on damaged fabric samples demonstrates detection accuracy of up to 96%, along with reduced processing time and operational cost compared to conventional fault detection techniques. Machine learning transforms business management by enabling data-driven decision making, optimizing operations, improving efficiency, and supporting scalable and adaptive organizational processes. The integration of enhancement, feature extraction, and clustering improves the reliability and consistency of defect identification while minimizing dependency on manual inspection processes. This framework supports scalable deployment in textile production environments and contributes to automated quality assessment by ensuring faster, more accurate, and cost-effective fault detection.
This study examines the digital consumer confidence-building process across the BRICS+ countries (2014-24) and addresses the critical gap in understanding how individual consumers develop digital platform confidence in emerging markets. We build on trust theory with an integrated socio-technical model and show that, through a consumer-centric analytical approach, the development of consumer trust in digital ecosystems can be explained by four behavioral stages (i.e., initial perception, experiential validation, relationship building, and ecosystem embeddedness) based on patterns in secondary data from multiple sources. Each stage demonstrates progressively stronger correlation coefficients (β = 0.74), with significant differences in consumer behavior patterns across stages. This research contributes to IS literature by examining how technological infrastructure relates to developing trust (R2=0.824) as well as institutional factors relating to forming trust (R 2 = 0.856). Although our secondary data sources do not allow us to directly measure the psychological processes, these patterns are consistent with previous work on risk perception and mediating roles of trust. In novel-to-consumer studies, we found significant variations in how consumers across different BRICS+ nations interpret identical trust signals, with cross-cultural dimensions accounting for 72.4% of this variance. Our segmentation analysis reveals three types of consumer trust archetypes: tech-confident adapters, cautious validators, and guided adopters, based on their different patterns of decision-making, risk thresholds, and platform interaction behaviors, which further extends the existing technology acceptance models to show the transition from technology acceptance to embedded behavior, and offers implications for digital platform designers and marketers in emerging economies, as well as advancing information systems understanding by mapping how consumers cognitively process technical, social, and institutional signals in the context of an evolving digital platform.
Climate change is a critical global challenge with significant implications for the environment, economy, and society. In response, governments worldwide have implemented various policies aimed at mitigating its effects. This article investigates why some countries are more innovative than others by examining the role of institutionalization of climate change mitigation policy on technological innovation. It also explores the mechanisms through which this institutionalization impacts innovation. Using cross-sectional data from 111 countries, the analysis employs various econometric techniques, including ordinary least squares regression and two-stage instrumental variable methods. The findings reveal that institutionalization of climate change mitigation policy has a positive and significant effect on technological innovation. Countries that are strongly committed to addressing climate change tend to foster greater innovation. Furthermore, mediation analysis shows that this institutionalization positively influences technological innovation indirectly by enhancing human skills, research and development expenditure, clean fuel adoption, and access to affordable energy.
The customer churn prediction is crucial for e-commerce business as it enables them to develop effective customer retention strategies and execute successful marketing operations. The existing works focus on the correlation rather than the causal factors, so it is difficult to inform the effective interventions. This paper introduces the LSTMRM approach for the effective customer churn prediction. First, the dataset of the customer behavior is collected. Next, preprocessing involves the imputation of missing values with sparsity handling using the KNTSN approach; then, de-duplication and data type conversion are done. After that, from the preprocessed results, dynamic customer segmentation is done by KH-PKC, and simultaneously feature extraction occurs. Then, the customer churn is defined by the CPH model; after that, CFGL is utilized based on the CPH output and the extracted features. Next, the trend and seasonal pattern are recognized by using the DWC-CHT. After that, DTW visualization is done from the results of the trend and seasonal pattern, and feature extraction takes place. Next, from the extracted features of DTW visualization, the churn is predicted by the LSTMRM approach. Finally, the predicted outcome from the LSTMRM approach and the churn-defined outcome from the CFGL model are stored in a cloud server. As per the experimental analysis, the proposed model attained 98.95% accuracy.
This study examines the conditions under which small and medium-sized enterprises (SMEs) decide to use government support programs for technological innovation. Using firm-level data from the 2018 Korean Innovation Survey (KIS), the analysis focuses on 1285 manufacturing SMEs. The results show that SMEs are more likely to use government support programs when competency-related barriers hinder innovation. Market-related barriers, including fierce competition and uncertain demand, also increase the likelihood of program participation. In contrast, financial constraints were negatively associated with the use of government support programs. Regarding R&D activities, SMEs conducting internal, joint, or external R&D were more inclined to use government support programs, highlighting the complementary relationship between innovation efforts and institutional support. These findings provide policy implications for designing differentiated support policies that strengthen competencies, encourage collaborative R&D, and mitigate market uncertainty. This study contributes to the understanding of SMEs’ innovation behavior and serves as a foundation for further research on innovation strategy and public policy.
Given the role of the Internet of Things (IoT) in achieving sustainable health, identifying and prioritizing IoT applications in the health field is a main challenge for policymakers, a challenge that has been addressed by many researchers. This paper presents a hybrid framework designed to address challenges and assist health policymakers in leveraging IoT technology effectively. The most important applications of IoT in hospitals have been identified through a systematic literature review and expert opinion. By combining the developed group BWM-G with four ranking methods (SAW-G, TOPSIS-G, ARAS-G, and EDAS-G), IoT applications in the hospital sector have been prioritized based on SD criteria and sub-criteria. Sensitivity analysis has been conducted to identify the most critical criterion and determine the robust ranking method. The proposed steps have been applied to a case study to illustrate how this framework can be used in practice. The research shows that hospital care and treatment are the most important factors in improving community health. Technologies such as real-time patient monitoring and smart resource management help hospitals operate sustainably and are ready for investment. To achieve sustainable development goals, the government should focus on these priorities and formulate appropriate policies to fully benefit from these technologies.
In recent years, there has been a notable rise in academic interest regarding studies on leadership styles and innovative workplace behaviors (IWB). Although its significance, there has been a lack of effort in evaluating the expanding research. This research utilizes the TCCM (Theory, Context, Characteristics, Methods) framework to elucidate the prevailing theories, contexts (such as industries and nations), characteristics (variables and their interrelations), and methodologies (encompassing research approaches and analytical techniques) applied in the study of IWB and leadership styles. Using PRISMA-guided methodologies, this study aims to thoroughly examine the literature regarding the effects of IWB and leadership styles by analyzing 91 studies to summarize and synthesize the current body of knowledge and delineate potential future studies. The review reveals a predominance of singular theoretical approaches, with research mostly focusing on theories such as Social Exchange and Social Cognitive Theories. The greater part of studies has been conducted in Asian countries, with China leading ([Formula: see text]), and the most significant industries are IT/technology/telecommunications, hospitality, and health. In this systematic review, independent, moderator, mediator and control variables were examined in detail by dividing them into categories such as behaviors, perceptions and intentions. This review synthesizes insights from prior research and proposes a future research agenda in underexplored contexts, such as industrial and national levels, by employing emerging theoretical frameworks and integrative analytical methods.
We examine the significance of leader traits as levers of morale and how team morale drives behavioral and success criteria in projects. Regression results utilizing 214 members in 65 projects illustrate leader agreeableness, extroversion, as well as openness, play a part in the development of morale. Additionally, results show morale has ramifications for success by boosting contextual performance behavior — when workers go the extra mile during project implementation. Implications for selecting and coaching personnel to spearhead projects are provided, which have consequences for morale as well as contextual performance behavior and project success.
As organizations have become increasingly reliant on technology and data to perform theirduties, the pressure on their executives has increased proportionally. Leaders must guide andmotivate their teams in order for their companies to meet the requirement of continuouslydeveloping new innovation and information frameworks. Therefore, the purpose of currentstudy is to examine the in degrees uence of participative leadership on employee innovative workbehavior through the mediating role of employee knowledge sharing attitude and absorptivecapacity in the context of project-based information technology (IT) organizations in Portugal.This study included 306 IT professionals working in di (R) erent public and private IT organizations.The statistical techniques employed for the analysis of collected data. IBM SPSS StatisticalVersion 23 and Smart PLS Version 4 were used for data analysis. The result revealed thatparticipatory leadership has a signi ficant and positive e (R) ect on employee innovative workbehavior. In addition, absorptive capability and employee knowledge sharing attitudes weresigni ficant mediators between participative leadership and employee innovative work behavior.The current recommendation is that leaders who are comprehensive, moral, engaging, attentive,and have a crystal-clear vision may inspire their followers to adopt novel approaches to their jobs.Lastly, we discussed future possibilities and the theoretical and practical implications of our findings.
This study examines the influence of sustainability dimensions, economic, environmental, and social, on tourist loyalty in the ecotourism sector, with tourist experience incorporated as a mediating variable. Tourist loyalty is crucial for the long-term viability of ecotourism, contributing to both environmental preservation and economic resilience. Data were collected through structured questionnaires at multiple ecotourism destinations, yielding 289 valid responses for analysis. Partial Least Squares Structural Equation Modelling (PLS-SEM) was employed to test the proposed relationships. The findings reveal that all three sustainability dimensions significantly enhance the tourist experience, which, in turn, positively drives tourist loyalty. While economic sustainability significantly affects the tourist experience, it does not directly influence loyalty, suggesting that experience is the key pathway through which economic factors translate into loyal behavior. Both environmental and social sustainability demonstrate direct and indirect effects on loyalty, highlighting their central role in shaping favorable visitor perceptions and behavioral intentions. The study suggests future research avenues, including longitudinal designs, qualitative insights, and the examination of potential moderating variables such as cultural background and tourist motivation. Practically, the results underscore the importance for ecotourism operators to adopt authentic sustainability practices and curate memorable experiences to cultivate loyal visitors. By strategically integrating sustainability initiatives with engaging experiences, destinations can foster repeat visits and positive word of mouth, thereby strengthening the long-term sustainability of the ecotourism sector.
This study develops an integrative green marketing model by examining the effects of green product innovation and digital marketing on sustainable marketing performance, with green brand image acting as a mediating variable among Generation Z consumers in Bali. A mixed-method sequential explanatory design was employed, combining quantitative analysis using Partial Least Squares-Structural Equation Modeling (PLS-SEM) with qualitative interviews to enrich the interpretation of the findings. The results reveal that green product innovation and digital marketing significantly enhance green brand image, which in turn positively influences sustainable marketing performance. While digital marketing also has a direct positive effect on performance, the impact of green product innovation is primarily indirect through green brand image. These findings highlight the critical role of brand image as a mechanism linking environmental innovation and marketing outcomes. Beyond marketing performance, the study demonstrates that green product innovation contributes to measurable environmental benefits through eco-design, recyclable materials, and energy-efficient production processes. These practices align with the Natural Resource-Based View (NRBV), emphasizing that environmental responsibility can serve as a source of long-term competitive advantage. This study provides both theoretical and practical insights, particularly for micro, small, and medium enterprises (MSMEs), by emphasizing the importance of integrating green innovation, digital strategy, and brand image management to achieve sustainable business growth in emerging markets.
The introduction of high-quality development has made enterprises face new challenges in development and business models. Enterprises actively respond through sustainability practices. Based on the dynamic capability theory, the study explores the influence of the dimension of the sustainability management control system (SMCS) and the high-quality development of enterprises. Using quantitative methods, survey data were collected from medium and large manufacturing enterprises in China. About 362 valid responses were analyzed through the partial least squares structural equation model (PLS-SEM). The study indicates that interactive systems have a positive promoting effect on the high-quality development of enterprises. However, diagnostic systems did not directly impact high-quality development due to strict performance metrics and management practices. Dynamic capabilities have demonstrated a mediating mechanism. Multi-group analysis (MGA) show that the research conclusions are not affected by the firm size and age. The research provides new insights for China's sustainability accounting. For enterprise managers, it is necessary to balance the diagnostic system and the interactive system to stimulate dynamic capabilities, so as to obtain new enterprise competitiveness and achieve sustainable development transformation.
Digitalized Innovation Environments (DIEs) are structured platforms that foster collaboration between universities and industry, providing advanced technological resources and a flexible space for innovation, skill development, and knowledge exchange. DIE is an umbrella term encompassing facilities such as makerspaces, FabLabs, and hackerspaces. This study examines the role of DIEs as incubators for industry-academia partnerships, with a focus on identifying the various collaboration formats established within these environments. The research identified three primary formats: short-term training initiatives, project-centered engagements, and long-term strategic alliances. Each format addresses different needs within the innovation ecosystem, ranging from skill acquisition and rapid prototyping to sustainable organizational transformation. These impacts were uncovered through a mixed-method case study approach, which followed and analyzed 47 corporate and startup collaborations over a seven-year period. Through a systematic process of data analysis and triangulation, additional factors were identified that play a significant role in the success of these collaborations. These factors are as multifaceted and complex as DIEs themselves, encompassing aspects such as government funding, administrative hurdles, and the diversity of the personnel involved. An accompanying survey of 39 collaboration participants further validated the qualitative findings, forming the basis for key practical implications. By providing a structured understanding of the functions and impacts of DIEs, this study offers valuable insights for optimizing these environments as dynamic centers of open and digital innovation and cross-sector collaboration, laying the groundwork for future research within university-industry ecosystems.
Corporate Social Responsibility (CSR) has evolved from a discretionary luxury to a strategic driver of organizational resilience and performance. This study examines how CSR initiatives, combined with disruptive innovation, influence firms' adaptive capacities and outcomes, while assessing the moderating roles of Research and Development (R&D) investment and competitive intensity. Analyzing survey data from 435 firms across 19 industries through structural equation modeling and contingency theory, the research reveals that proactive CSR engagement strengthens organizational resilience during crises, enhancing performance metrics such as return on assets and sales growth. Disruptive innovation mediates this relationship, with R&D intensity amplifying CSR's positive effects, particularly in highly competitive environments. Both young and mature firms demonstrate accelerated recovery and sustained performance when integrating CSR with innovation strategies. The findings underscore CSR's dual function as a risk mitigator and value creator, offering managers and policymakers actionable insights for aligning CSR investments with resilience-building priorities. This study advances stakeholder and contingency theories by positioning CSR as a critical lever for sustainable economic development in volatile markets.
This study investigates how university-industry (U-I) collaboration contributes to firms' innovation capabilities and innovation performance, focusing on two distinct modes of interaction: (i) universities as platforms for knowledge and innovation; and (ii) universities as providers of technological solutions. By distinguishing these modes, the paper contributes to a broader understanding of how heterogeneous university resources influence firms' ability to innovate. Using survey data from 176 Brazilian manufacturing firms engaged in collaboration with universities, the analysis reveals that interactions involving universities' knowledge resources - such as access to research outputs, laboratories, and talent - are more beneficial for enhancing firms' innovation capabilities than those based on direct technological solutions. These findings offer a holistic view of the benefits associated with different U-I collaboration strategies and highlight the relevance of university knowledge platforms for firms seeking to build long-term innovation capability.
