In this article, practical issues are considered at the complex level, in the field of realizing mathematical modelling of the assessment of the potential of the Ukrainian IT industry in the conditions of global challenges. The purpose of the study is to analyze modern approaches to evaluating this potential using advanced mathematical methods, such as multifactorial regression, cluster analysis and factor analysis. The work describes in detail the methods that allow you to evaluate the relationships between the main economic, technological and social indicators that influence the globalized development of the modern IT sector in Ukraine. The study emphasizes the strategic importance of assessing the capabilities of the IT sector to solve modern economic and technological tasks. Within mathematical modeling in the work, an improved model of evaluation of the above potential is proposed in which the combination of considered methods with the combination of the principles of Bayesian data analysis is combined. This approach allows you to more accurately take into account the uncertainty and variability inherent in modern economic and technological conditions. Usage of the Bayesian approach makes it possible to get a more flexible and adaptive model that better reflects dynamic processes that affect the development of IT industry in Ukraine in global challenges. The analysis begins with multifactorial regression, which examines the relationship between economic, technological and infrastructure factors that influence the development of the IT sector. The next stage is a cluster analysis that allows you to distinguish regional IT-hubs, comparing their effectiveness and revealing differences in Ukraine. With the help of factor analysis, we reveal the hidden variables that have a significant impact on the development of the industry. We also compare the results of each of the models to identify common trends and differences, which allows not only to identify opportunities for development, but also to indicate problems that need attention. Based on the results obtained, we offer specific recommendations for improving the competitiveness and stability of the Ukrainian IT industry. Our conclusions emphasize the importance of using innovative approaches to achieve more efficient development of this field.

Agricultural price prediction in developing regions faces significant challenges from missing data in Internet of Things (IoT)-based environmental monitoring systems, particularly in tropical fruit cultivation where sensors frequently experience connectivity and operational failures. This study evaluates the impact of missing data imputation methods on agricultural price prediction model performance using environmental and market data from a commercial durian orchard in Chanthaburi Province, Thailand (2023-2024). Three imputation strategies—Linear Interpolation, Prophet, and Kalman Filter—were systematically compared across four machine learning algorithms (Regression Trees, Random Forest, XGBoost, and Artificial Neural Networks) using 10-fold cross-validation. The dataset comprised 182 observations with 28.02% missing environmental data and 68.13% missing price data, representing realistic constraints in developing agricultural economies. Results demonstrated that XGBoost consistently achieved superior performance across all imputation methods, with Kalman Filter combined with XGBoost showing the best testing performance (R² = 0.9767, MSE = 0.0013, MAE = 0.0287, MAPE = 1.49%). However, these results require careful interpretation given the limited sample size, high missingness, and potential temporal data leakage from random train-test splitting. Time series visualization revealed distinct characteristics: Linear Interpolation provided computational efficiency but oversimplified data complexity, Prophet captured seasonal patterns but introduced excessive noise, while Kalman Filter offered balanced performance preserving both smoothness and natural variability. Practical price prediction analysis showed substantial variations up to 35 Thai Baht per kilogram between imputation methods. The findings provide methodological evidence for imputation strategy selection in agricultural IoT systems with missing data, though validation with larger multi-site datasets is essential before operational deployment.

With increasing developments in artificial intelligence and the need for more personalized digital experiences, user trust and engagement have become relevant factors to be considered for the success of e-commerce recommender systems. This study presents a bibliometric analysis of research trends from 2003 to 2023 by exploring the evolution of trust and engagement in this domain. Using data from the Scopus database, we investigated publication trends, influential works, key contributors, and emerging research themes. Our results reveal a surge in research output between 2020 and 2023, which shows an increasing scholarly appreciation of trust as a critical determinant of user engagement of recommender systems. The leading role of China in global contributions emphasized its reliance on social commerce models, where recommendations are powered by a community-based trust mechanism to drive user engagement. While foundational topics such as collaborative filtering and machine learning remain central, emerging themes (explainability, blockchain integration, and adaptive AI) highlight a shift toward more user-centric and secure systems. These reinforce trust through transparency and security while boosting engagement through active personalization. Thematic evolution from algorithmic development to AI-driven innovations shows how transparency, personalization, and security serve as vital trust-building influencers that drive user engagement in recommender systems. Also, regional disparities in research output, especially in Africa and South America reveal considerable gaps in understanding culturally specific trust factors and engagement patterns. This indicates the need for collaborative studies to develop inclusive recommender systems tailored to local context to bridge these gaps. These findings reflect that trust and engagement are not simply complementary features, but fundamental pillars that are influencing the future of e-commerce recommender systems. As AI advances toward explainable, secure, and adaptive designs, this research calls for urgent globally inclusive frameworks that address both technological sophistication and cultural diversity to ensure that recommender systems emerge as equitable tools for global e-commerce.

The rapid growth of mobile wallet usage has led to a sharp increase in fraudulent transactions, making fraud detection in portable wallets a pressing concern. Accurately detecting fraud is difficult because transaction data is complicated and unbalanced. Conventional rule-based systems are less flexible and frequently provide large false positive rates along with poor accuracy. Effective feature selection is crucial to the performance of Machine Learning (ML) models, notwithstanding their increased detection rates. Redundancy and noise are introduced by high-dimensional data, which lowers model performance and raises computing costs. The advantages of hybrid feature selection are frequently overlooked in current research, particularly when it comes to portable wallet fraud detection. By combining Random Forest Importance, LASSO Regression, Recursive Feature Elimination (RFE), and Mutual Information (MI) with resampling to solve class imbalance, this study fills that gap. Our approach provides a more reliable and effective solution for safe portable wallet fraud detection by removing superfluous features, increasing accuracy, and reducing computing cost. The model becomes faster and more effective when superfluous characteristics are eliminated because this reduces the computational effort. By concentrating just on the most instructive data, it increases accuracy. By addressing class imbalance and combining several selection strategies, the hybrid approach guarantees robustness. All things considered, this leads to a scalable and safe fraud detection system for transactions using mobile wallets. Our results show that a successful feature selection approach improves fraud detection accuracy, which in turn improves operational effectiveness and financial security. 

In modern educational environments, particularly within computer laboratory settings in higher education institutions, the lack of effective real-time supervision and streamlined assessment processes presents a persistent challenge. Most current systems still rely on manual monitoring and evaluation, which are not only inefficient and time-intensive but also vulnerable to academic dishonesty, such as copy-paste behaviour during lab work. This study identifies and addresses this critical gap by proposing the development and implementation of an integrated real-time monitoring and assessment system tailored for use in academic computer labs. The proposed solution is a desktop-based application that incorporates four key features: Real-Time Viewer (RTV) for live monitoring of student activities, Block Inappropriate Websites (BIW) to restrict access to non-educational or harmful content, Manage Computer Time (MCT) to regulate system usage duration, and Form Learning Assessment (FLA) for digitalized and efficient performance evaluation. The development process followed the System Development Life Cycle (SDLC) framework, ensuring a structured approach across analysis, design, implementation, testing, and maintenance stages. Empirical testing involved a series of functional test cases simulating real-use conditions. All seven critical scenarios—such as input validation, session management, access control, and data deletion—were executed and passed successfully, indicating the system’s robustness and usability. In a pilot study conducted at Pekanbaru College of Technology, the application was tested among 30 students across multiple laboratory sessions. The results demonstrated a notable improvement in student engagement and learning performance. Quantitatively, students achieved learning assessment scores ranging from 84 to 96, with a calculated mean of 89.6 and a standard deviation of 4.1. These outcomes suggest that the introduction of automated, real-time monitoring significantly enhances not only instructional supervision but also the accuracy and fairness of learning assessments. This research contributes to the field by bridging the gap between digital classroom management and performance assessment in a higher education context. It introduces an innovative and practical approach for educators to maintain instructional quality while managing multiple learners in digital settings. Moreover, the findings provide empirical evidence supporting the integration of real-time supervision tools into educational systems to foster accountability, deter academic misconduct, and support data-driven instructional improvements.

Employee attrition is an important factor that can affect organizations, both financially and operationally. Human Resource (HR) managers often find it difficult to identify exactly which employees might be planning to leave the organization and what is the root cause for their decision. With the recent advances in computing, Machine Learning (ML) techniques are available for analysing, understanding, and solving complex problems. This study analyses the IBM HR Analytics dataset using ML techniques to predict employee attrition and identify the key factors that influence attrition. Four ML models based on Logistic Regression, Decision Tree, Random Forest, and Gradient Boosting have been used for analysing attrition. It is found that Logistic Regression outperformed all other models in predicting attrition. At the same time, Decision Tree is found to be the weakest among the four techniques. On the analysis of feature importance, it is found that variables related to compensation (Monthly Income), career stage (Total Working Years, Age), and tenure at the organization are among the most significant factors influencing attrition.  The insight from this study is expected to help HR managers in developing effective, data-driven strategies to retain their talent in their organization.

Despite cloud computing's scalability and economy, energy efficiency, security, and equitable scheduling remain significant concerns. The traditional scheduling approach often fails to optimize execution time, energy consumption, and security concerns, resulting in less resource utilization and less secure systems. This paper proposes the Hybrid Bat-Genetic Algorithm (HBA-GA), which combines the Bat Algorithm for fast exploration with the Genetic Algorithm for accurate exploitation. This method reduces energy use while also reducing security risks like unauthorized access and data leaks. It uses Jain's Fairness Index (JFI) in order to ensure that workloads are evenly distributed and VM overload and conflicts are avoided. Based on simulations results, proposed HBA-GA improves energy efficiency while reducing security exposure and risk likelihood at the scheduling level by incorporating security-aware risk scoring into task–VM allocation decisions.

In the dairy industry, optimizing reproductive management is crucial for sustainable operations and enhancing animal welfare. The traditional manual detection methods usually miss many of the estrus incidences and hence have resulted in a 20-30% decline in conception rates and further massive economic losses.This paper presents an advanced framework integrating machine learning and Internet of Things (IoT) technologies to improve estruses detection in dairy cattle, thereby supporting efficient herd management and productivity. The proposed solution leverages a stacking model of Random Forest and Gradient Boosting Machine (GBM) algorithms to accurately identify estruses events, providing a reliable method for reproductive monitoring. The experimental evaluation yields accuracies of 92.1 % using RF, 92.3 % using GBM, and an improved 93.19 % when the stacking model is applied, along with improvements in precision of 94 and an F1-score of 94 %, reflecting its strength in complex behavioral pattern recognition. Rigorous evaluation across key performance metrics confirms the model’s high accuracy, underscoring its suitability for practical deployment. The system employs IoT-enabled smart collars equipped with temperature sensors, accelerometers, GPS, and RFID to gather real-time data on cattle health and reproductive status. By analyzing this data, the system delivers precise and timely insights into estruses cycles, enabling targeted breeding interventions and enhanced reproductive management. Data collected through the smart collars is securely stored in Google Firebase, facilitating efficient data archiving and rapid access via a user-friendly web application. The proposed integration of IoT, machine learning, and cloud computing presents a holistic, scalable, and economically viable solution for enhancing reproductive efficiency, animal welfare, and sustainable dairy management.

In recent years, the rapid advancement of machine learning (ML) has surpassed many expectations, and its application in the healthcare sector has emerged as one of the most fascinating areas of exploration. This thesis looks into whether machine learning can increase the precision and efficacy of breast cancer diagnosis. With the help of nine classification algorithms including Random Forest, XGBoost and MLP Classifier the given work intends to propose a reliable automatic solution for malignant and benign classification of breast tumor. The main idea of the project is the development of the Web based tool that would allow doctors and other medical practitioners to make quick decisions The MLP Classifier was found to be the optimal solution after its efficiency was evaluated based on the accuracy rate, and such parameters as precision rate, recall rate, and F1-score. This leads to development of a user friendly app; even those that would not originally consider themselves technical can easily operate the application. Apart from addressing the matter of high accuracy of diagnostics, the system shows the possibility of minimizing the rates of human factors and optimizing clinical decision. Seeking for that day when technology and human opinion will complement each other in the delivery of healthcare, our study neither only contributes to the growing literature on applying artificial intelligence in healthcare but also evolves the blueprint to integrate ML models in everyday practice.

The rise of FinTech lending in India has transformed credit access, yet studies examining customer experiences with artificial intelligence (AI)-enabled FinTech lending platforms remain limited. This study investigates the key drivers of user experience and the evolving sentiment toward AI-enabled lending platforms by analysing online reviews from 2017 to 2024 using LDA topic modelling and lexicon-based longitudinal sentiment analysis. Twelve key topics emerged, revealing significant negative sentiment around customer support, eligibility checks, documentation, repayment, and app trustworthiness. In contrast, app usability and interface design maintained strong positivity, while loan approval and disbursement processes saw declining sentiment. Despite these pain points, overall user experience remained positive, indicating that the perceived benefits such as speed, efficiency, and convenience provided by these platforms outweighed concerns like high interest rates, privacy risks, and poor customer service. The findings highlight a nuanced balance between technological advantages and operational shortcomings, offering insights for improving AI-enabled lending platforms.

The beginning of the fourth industrial revolution or Industry 4.0 has changed the concept of automation in industries by adopting the Internet of Things (IoT) in manufacturing, logistics, and production processes. The IoT is the digital foundation of Industry 4.0 that allows real-time monitoring, predictive maintenance, data-based decision making, and autonomous processes with in-between devices and smart sensors. This review explores the uses of the IoT in industrial automation through analyzing the enabling technologies, communication protocols, integration with cloud computing, wireless sensor networks, edge computing, artificial intelligence (AI) and machine learning (ML), as well as presenting important applications, current challenges, and future trends in smart industrial systems. Instead of considering the technologies separately, this paper takes the system-level viewpoint by integrating the way IoT architecture, communication protocol, intelligent analytics, and security controls collectively facilitate Industry 4.0 automation.

Diabetes mellitus is a chronic metabolic disorder with a rapidly increasing global prevalence, posing a significant public health challenge. Early detection of diabetes can enable timely intervention and preventive measures, thereby reducing the risk of long-term complications. In this study, a machine learning (ML)-based methodology is proposed for the early prediction of diabetes mellitus. The proposed approach enhances existing prediction systems by improving key performance metrics, including precision, recall, and F1-score, and achieves an efficiency improvement of 4%–10% compared to state-of-the-art methods. Experimental results demonstrate that the support vector machine outperforms other ML algorithms for diabetes prediction, achieving 92% accuracy, 95% precision, 92% recall, 93% F1-score, 92% specificity, and an area under the receiver operating characteristic curve of 0.97.