Bulletin of Electrical Engineering and Informatics
Not a member yet
2885 research outputs found
Sort by
Enhancing recommendation diversity in e-commerce using siamese network and cluster-based technique
Get PDFThis study investigates the difficulty of improving product recommendations in e-commerce systems by tackling the common problem of poor diversity in suggestions. We present a novel approach that uses a siamese network architecture and ResNet for feature extraction to recommend visually similar elements while incorporating diversity through a cluster-based mechanism. The Siamese network is used to compare product pairs, allowing it to recommend both comparable and dissimilar items from distinct clusters. The model was evaluated using a variety of evaluation metrics, resulting in an accuracy of 88.5%, a precision of 90.2%, a recall of 87.1%, and an F1 score of 88.6%. Our results demonstrate that our strategy maintains a high level of relevance in suggestions while efficiently incorporating variety, hence improving the overall user experience in e-commerce applications
Enhancing SDN security with a feature-based approach using multiple k-means, Word2Vec, and neural network
Get PDFIn the rapidly evolving landscape of network management, software-defined networking (SDN) stands out as a transformative technology. It revolutionizes network management by decoupling the control and data planes, enhancing both flexibility and operational efficiency. However, this separation introduces significant security challenges, such as data interception, manipulation, and unauthorized access. To address these issues, this paper investigates the application of advanced clustering and classification algorithms for anomaly detection and traffic analysis in SDN environments. We present a novel approach that integrates multiple k-means clustering models with Word2Vec for feature extraction, followed by classification using a neural network (NN). Our method is rigorously benchmarked against a traditional NN model to comprehensively evaluate performance. Experimental results indicate that our approach outperforms the NN model, achieving an accuracy of 99.97% on the InSDN dataset and 98.65% on the CIC-DDoS2019 dataset, showcasing its effectiveness in detecting anomalies without relying on feature selection. These findings suggest that integrating clustering techniques with feature extraction algorithms can significantly enhance the security of SDN infrastructures
Refining CNN architecture for forest fire detection: improving accuracy through efficient hyperparameter tuning
Get PDFForest fire detection is one of the critical challenges in disaster mitigation and environmental management. This research aims to increase the accuracy of forest fire detection through improving the convolutional neural network (CNN) architecture. The main focus of research is on efficient hyperparameter tuning, which includes selecting and optimizing key parameters in CNN architectures such as convolutional layers, kernel size, number of neurons in hidden layers, and learning algorithms. By utilizing grid search techniques and heuristic-based optimization algorithms, the resulting CNN model shows significant improvements in detection accuracy compared to previous approaches. The evaluation was carried out using a pre-processed forest fire image dataset, and the results show that architectural refinement and appropriate hyperparameter tuning can substantially improve model performance. Evaluation results comparing two models, VGG16 and the proposed method, show significant improvements over the proposed method. The proposed method shows better capabilities with an accuracy of 95.31% and a precision of 97.22%. This research contributes to developing a more reliable and efficient forest fire detection system, which is expected to be used in real applications to reduce the impact of forest fires more effectively
Design and analysis of an asymmetrical star-shaped fractal antenna with meta-surface integration at 5.2 GHz
Get PDFWireless communication requires optimized antenna designs to ensure maxi mum signal reception and transmission in today’s rapidly advancing technolo gies. Recent research emphasizes improving antenna efficiency and directivity to support higher data rates, extended coverage, and reliable connectivity. How ever, conventional antenna structures often suffer from narrow bandwidth, low radiation efficiency, and high return loss, which degrade signal quality and re strict operational range, particularly in complex electromagnetic environments. This study introduces an innovative asymmetrical star-shaped fractal antenna coupled with a metasurface layer consisting of periodic split-ring resonator (SRR) unit cells on a FR4 substrate to overcome these restrictions. The SRR based metasurface plays a critical role in suppressing surface waves, improving impedance matching, and enhancing radiation directivity. Experimental evalu ations were performed across 4.5–10 GHz, focusing on key performance mea sures such as gain, return loss, and voltage standing wave ratio (VSWR). The suggeted antenna achieved a stable return loss below −10 dB and demonstrated a strong operational peak at 5.2 GHz, with improved directivity and radiation efficiency compared to conventional patch designs. The integration of asym metrical star-shaped fractal geometry with SRR-based metasurface technology effectively addresses the shortcomings of traditional antennas, establishing the proposed design as a compact, efficient, and reliable candidate for mid-band wireless communication systems
Design and emulation of an SDN network with opendaylight to improve QoS in a peruvian financial institution
Get PDFThis study presents the design and emulation of a software-defined networking (SDN) architecture using the OpenDaylight controller to enhance the quality of service (QoS) in a Peruvian financial institution. The main objective is to overcome limitations of traditional networks, including high latency, limited bandwidth, and packet loss, which hinder the efficiency of financial services. The proposed SDN architecture was implemented and tested through simulations in the Eve-NG platform, where key performance parameters—latency, throughput, and packet loss—were measured. Results demonstrated significant improvements, with latency reduced by up to 40%, stable throughput maintained at 10 Mbps across all branches, and a noticeable reduction in packet loss. These outcomes validate the feasibility of adopting SDN in financial environments to support critical services and ensure operational continuity. Furthermore, the findings emphasize SDN’s role in modernizing network infrastructures, improving user experience, and aligning local financial institutions with international technological trends. Future research may explore alternative SDN controllers, scalability in larger topologies, and integration with emerging technologies such as network function virtualization (NFV)
Thermal analysis of li-ion battery pack using phase change materials based on climate conditions
Get PDFThe development of lithium-ion batteries necessitates improved management of these systems, particularly with regard to thermal aspects. They operate optimally between 35 °C and 45 °C. Temperatures exceeding 50 °C accelerate cell aging, while those surpassing 60 °C can trigger thermal runaway, potentially leading to catastrophic failure. To mitigate these risks, phase change materials (PCMs) are employed in battery thermal management systems (BTMS). They absorb heat during charging or discharging, transitioning from solid to liquid, then release the stored energy during periods of low demand, solidifying to help regulate battery temperature. This study conducts a thermal analysis of a lithium-ion (LiFePO4) battery pack delivering a 24 V load, using COMSOL MULTIPHYSICS software. The objective is to evaluate and compare the thermal behavior of different PCMs, RT27, Paraffin Wax 58-60, and HM030, against air as a baseline reference. Simulations are performed using the integrated finite element method (FEM), with a discharge rate of 4 C. A correlation is proposed between the choice of PCM and the climate in specific locations, with the choice being made based on the disparities in the results obtained
Evaluating the impact of risk management and cybersecurity on decision-making in the Peruvian National Informatics System
Get PDFThis research addresses the influence of risk management (RM) and cybersecurity (CIB) on decision making (DM) of the Peruvian State's National Informatics System (SNIEP) in the year 2024, in line with sustainable development goal 9. Using a quantitative, non-experimental, cross-sectional design approach, a sample of 487 CIB analysts was analyzed to explore the relationship between these critical variables. The findings show uneven implementation of RM and CIB practices, which significantly impact the DM and quality of information (Sig 0.000), processes (Sig 0.001), and the effectiveness of system decisions (Sig 0.060). In addition, key areas were identified to strengthen the integration of RM and CIB strategies in the state's digital environment, highlighting their importance to ensure informed and resilient decisions in the face of growing cyber threats. The study provides empirical evidence on their impact on the quality, effectiveness and security of DM in government digital environments. The research contributes both to the development of a theoretical framework that articulates concepts of RM, CIB, and DM in the public sector, and to the formulation of strategies and policies that promote a secure and efficient digital infrastructure, aimed at improving public services and citizen trust in the contemporary digital environment
Explainable artificial intelligence for multiclass prediction model of suicide attempt
Get PDFSuicide attempt prediction is a challenging classification problem that involves a variety of risk factors in individuals with various medical conditions. Accurate risk stratification prediction is hampered by the absence of reasons for those who have attempted suicide and developing prediction model is challenged to be explained. Therefore, this work aimed to develop a multiclass prediction model for suicide attempts and to use Shapley additive explanations (SHAP), an explainable artificial intelligence (XAI) method to analyze the prediction model for suicide attempts in explaining the decision of the model. The prediction model is trained using machine learning approaches, random forest (RF) and gradient boosting (GB), on a clinical dataset of patients with chronic diseases. GB demonstrated higher accuracy with 0.81 than RF with 0.78 for multiclass classification results (no risk, low risk, moderate risk, and high risk). By analyzing the SHAP explanations, clinicians can gain valuable insights into the factors contributing to suicide attempt predictions in patients with chronic diseases. This enhanced understanding can facilitate more informed and appropriate treatment decisions, potentially leading to improved patient outcomes and targeted interventions
Hybrid DL and ML approach for MRI-based classification of bone marrow changes in lumbar vertebrae
Get PDFAlterations in the bone marrow changes lumbar vertebrae (BMCLVB) are considered important markers of chronic low back pain severity, particularly among patients with coexisting conditions like osteoporosis or cancer. Realizing these associations informs healthcare and insurance frameworks but also supports early intrusion planning for high-risk populations. This study aims to classification (BMCLVB) as normal or abnormal used magnetic resonance imaging (MRI) with machine learning (ML) model. A novel dataset comprising 1,018 BMCLVB MRI images was utilized to extract deep features via a pre-trained ConvNeXt-XLarge model. These features were then classified using different types in individual and ensemble ML algorithms. To ensure a comprehensive performance evaluation, all models were tested using accuracy, precision, recall, and F1-score. The combination of ConvNeXt-XLarge and logistic regression (LR) achieved a classification accuracy 93.14%, precision 93.22%, recall 94.83%, and F1-score 94.02%. These results highlight that the proposed model provides a fast and cost-efficient solution supporting the diagnosis of BMCLVB and potential to significantly improve clinical decision-making and patient care outcomes
A rigorous examination of electromyography forearm muscle response in grasping and swinging scenarios
Get PDFThis study examines the use of electromyography (EMG) in analyzing forearm muscle responses in hand grasping force with swinging motions. We start by establishing the basics of hand grasping force and swinging motions, laying the groundwork for subsequent discussions. The paper critically assesses various EMG techniques, highlighting how they reveal muscle activity during hand grasping in dynamic situations. We explore how swinging motions affect hand grasping force biomechanics, emphasizing the role of EMG in capturing dynamic muscle activity. A thorough examination of methodologies used in EMG studies provides insights into current practices and emerging trends. Practical applications across fields like rehabilitation and robotics underscore the relevance of this research. The study concludes by addressing current challenges and suggesting future research directions. This synthesis provides a straightforward resource for researchers, practitioners, and technologists seeking a deeper understanding of EMG indices in hand-grasping force analysis with swinging action