Vol. 6 No. 11 (2024): Volume 06 Issue 11

In the digital banking landscape, the increasing volume of online transactions has heightened the risk of fraudulent activities, necessitating the development of more effective detection systems. This study investigates the efficacy of various machine learning and deep learning algorithms in identifying fraudulent transactions, emphasizing Long Short-Term Memory (LSTM) networks. We implemented and evaluated multiple algorithms, including Logistic Regression, Random Forest, Gradient Boosting Machines (GBM), and XGBoost, on a large-scale credit card transaction dataset. Our results demonstrate that the LSTM model outperforms traditional machine learning algorithms, achieving an accuracy of 98.5%, precision of 87.2%, recall of 85.0%, and an Area Under the Curve (AUC) score of 0.94. These findings highlight the superior capability of LSTM networks to capture complex patterns in sequential transaction data, making them an asset for real-time fraud detection in banking. This research underscores the need for financial institutions to adopt advanced deep learning techniques to enhance their fraud detection systems, thereby minimizing financial losses and improving customer trust.

This article analyzes the effectiveness of a new incubator design for reviving silkworm eggs. In the course of the study, an incubator based on a rotating disk was created to control the parameters of the microclimate and ensure uniform development of eggs. In this system, temperature, humidity and CO2 levels are monitored using the SCD 41 and other sensors. The results showed that using the new incubator, silkworm eggs were revived 4.1% faster than using traditional methods, and the cocoon yield increased by 5.8%. Thanks to the high efficiency of the new construction, the economic efficiency of silk production will increase and the possibility of producing high-quality silk will expand. Thus, the construction of an innovative incubator opens up opportunities for using new approaches in sericulture.

In this study, a mechatronic system was developed for the care of mulberry silkworms in multi-layered succulents. This system allows you to monitor and monitor the humidity and temperature in each layer using the DHT11 sensor. The results showed that humidity and temperature were kept stable by a mechatronic system, which positively affected the growth of silkworms and the production of high-quality silk. Compared to traditional methods, the system has made it possible to use resources more efficiently and save energy. In short, the mechatronic system makes the silkworm rearing process efficient and stable. This suggests that the method has great prospects in the development of the silk industry.

This study investigates the design and efficiency of skimmer technologies in enhancing the functionality of active grease traps. Grease traps play a critical role in wastewater management by intercepting fats, oils, and greases (FOGs) from kitchen wastewater, thereby preventing clogging and operational issues in sewer systems. However, conventional skimming methods often face challenges related to efficiency and effectiveness. This research evaluates various skimmer designs, focusing on their operational efficiency, removal rates of FOGs, and overall impact on grease trap performance. Through empirical testing and comparative analysis, the study identifies key design features that optimize skimmer functionality. The findings suggest that advanced skimmer technologies can significantly improve grease trap efficiency, leading to better maintenance practices and reduced environmental impact. This research contributes to the development of more effective wastewater management strategies in commercial kitchens and similar settings.

The adaptation of system engineering to evaluate technical startups through an innovative framework based on the OMG Essence standard is a modern approach to simplify the process of analyzing and managing startups at various stages of their development. In this study, a universal framework was proposed that allows evaluating technical startups from the point of view of system engineering. This approach takes into account key aspects of startup development, such as requirements, stakeholders, technology, and team, which makes it an important tool for evaluating innovative projects.

The main purpose of the proposed framework is to apply the principles of systems engineering to the evaluation of startups, focusing on technical aspects such as system architecture, integration capabilities, and the ability of the team to solve complex tasks. Traditional methods of evaluating startups, often focusing on business aspects, may not always take into account all the technical difficulties and risks that arise when developing software products. The OMG Essence-based framework fills this gap by providing a more comprehensive tool for analyzing and managing startup development.

OMG Essence, as a standardized method, acts as a basis for formalizing practices and facilitating interaction between various project participants. The framework based on it offers the possibility of modular adaptation, which allows you to evaluate startups regardless of their complexity and current stage of development. An important feature of OMG Essence is the ability to integrate with other methodologies, which makes it universal for various fields. The application of this standard in the evaluation of startups allows you to obtain more objective results and improve the decision-making process.

The article presents a study of modern methods of training and adaptation of artificial agents used in managing complex processes, which are characterized by a high level of uncertainty and the need for prompt response to changes. Key methodological approaches such as machine learning and neuroevolution are discussed. These approaches allow AI agents to accumulate knowledge about the behavior of systems continuously, analyze external changes, and adjust the management strategy depending on environmental conditions, which significantly increases their ability to predict and prevent possible failures in management.

In the course of the study, models were considered that allow automating the execution of complex, multitasking processes, minimizing human intervention, and reducing the likelihood of errors. In addition, the presented methods provide high flexibility and scalability of systems, which is especially important in industrial and technological industries, where stability and reliability are critical. The results showed that AI agents with adaptive learning capabilities can increase operational efficiency while reducing costs and optimizing resource use. The conclusion highlights the prospects of using artificial intelligence to build highly autonomous control systems capable of responding to dynamic challenges, which opens up new horizons for automation and intellectual support in industrial production, logistics, and other key areas.

Thus, the article makes a significant contribution to understanding the role of AI in management modernization, offering practical recommendations on the implementation of intelligent agents in real-world scenarios to increase productivity and sustainability.

Rapid technological advancements have driven the increasing complexity of warfare, positioning unmanned aerial vehicles (UAVs), or drones, at the forefront of military operations globally. This study examines Nigeria's adoption of drone technology in its fight against insurgencies, banditry, and communal violence. As part of its evolving security strategy, Nigeria has integrated drones into its military operations to enhance intelligence gathering, surveillance, and precision strikes. Despite the operational benefits, the use of drones raises ethical concerns, especially related to civilian casualties, and poses challenges due to limited transparency and regulatory oversight. The study employed a systematic literature review (SLR), drawing on secondary data sources and reports that address the role, effectiveness, and challenges of drones in Nigeria's defense strategies. Key findings highlight that Nigeria's military, particularly the Air Force, has successfully deployed both indigenous and foreign-acquired drones, such as the Amebo, Gulma, and Chinese CH-4, to bolster its intelligence and combat operations. However, the lack of a robust legal framework and the need for improved training and data processing infrastructure hinder the full realization of drone technology's potential. Based on the findings, the study recommends enhancing Nigeria's regulatory framework to ensure ethical drone deployment, investing in indigenous UAV development, and improving inter-agency coordination for more efficient use of drone capabilities. These steps are crucial to ensuring that drones can effectively contribute to Nigeria's national security while minimizing the associated risks.

This study investigates the effectiveness of generative models and traditional classification models in detecting fraud and anomalies within the retail banking sector. Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs) were evaluated for their capability to generate realistic synthetic transaction data and identify anomalies, achieving anomaly detection accuracies of 91.2% and 93.5%, respectively. These models were also assessed using Inception Score and Fréchet Inception Distance (FID), with GANs exhibiting superior data realism. Among classification models, Gradient Boosting Machines (GBM) demonstrated the best performance, achieving an accuracy of 96.3%, a precision of 93.5%, a recall of 91.4%, and an AUC-ROC of 97.2%. Random Forest and Logistic Regression also performed well, though with slightly lower metrics.

This study presents a comprehensive analysis of machine learning techniques for measuring and predicting security in stock markets, comparing the performance of supervised, unsupervised, and deep learning models. Using a diverse dataset from Kaggle that includes historical stock prices, financial news sentiment, company fundamentals, and macroeconomic indicators, we applied feature engineering and rigorous preprocessing methods to optimize model accuracy. The study evaluated Random Forest, Support Vector Machines (SVM), K-Means clustering, and Long Short-Term Memory (LSTM) networks across key performance metrics. Results indicate that Random Forest outperformed other models in classification tasks with an accuracy of 92%, making it highly effective for real-time security assessment. SVM also demonstrated strong classification capabilities, particularly in high-dimensional spaces, with an accuracy of 88%. K-Means and DBSCAN clustering algorithms excelled in anomaly detection, identifying unusual patterns that could signal market irregularities. LSTM models, designed for time-series forecasting, achieved a root mean square error (RMSE) of 1.78, proving their utility in predicting future stock trends but requiring more computational resources.Our findings suggest that a hybrid approach, combining the strengths of supervised and deep learning models, can provide a robust solution for stock market security measurement. By leveraging explainable AI techniques such as SHAP and LIME, we also improved model interpretability, making these predictions more actionable for stakeholders. This research highlights the potential of machine learning in financial security monitoring and supports the growing integration of AI in the finance industry.