International Journal of Electrical and Computer Engineering (IJECE)
Not a member yet
6355 research outputs found
Sort by
Generalization of reactive power definition for periodical waveforms
Get PDFThe article presents a selection of reactive power definitions, which are applicable for implementation in energy meters. For sinusoidal current and voltage waveforms, all provided dependencies yield equivalent reactive power values. However, in the presence of distorted current and voltage, the power values are determined by the applied method (algorithm). Standardization requirements for reactive energy meters stipulate metrological verification under sinusoidal conditions. The selection of an optimal reactive power definition remains a problematic and ongoing subject of debate within the field. The paper shows that a generalized unique definition of additive reactive power derives from the definition of active power. Unlike active power, reactive power must be independent of the conversion of electric energy into work and heat. This independence is achieved if one of the waveforms – the current in the scalar voltage and current product (specifying active power) – is replaced by a special orthogonal waveform. An orthogonal waveform can be derived through either differentiation or integration. Reactive power obtained by this method is an additive within the system. When differentiation is employed, the reactive power for a nonlinear resistive load with a unique, time-invariant current-voltage characteristic will be zero. Some other properties of reactive power defined in this way are presented. This method is straightforward to implement in reactive energy meters
Students performance clustering for future personalized in learning virtual reality
Get PDFThis study investigates five clustering algorithms—K-Means, Gaussian mixture model (GMM), hierarchical clustering (HC), k-medoids, and spectral clustering—applied to student performance in mathematics, reading, and writing to support the development of virtual reality (VR)-based adaptive learning systems. Cluster quality was assessed using Davies-Bouldin and Calinski-Harabasz indices. Spectral clustering achieved the best results (DBI = 0.75, CHI = 1322), followed by K-Means (DBI = 0.79, CHI = 1398), while HC demonstrated superior robustness to outliers. Three distinct student profiles—beginner, intermediate, and advanced—emerged, enabling targeted adaptive interventions. Supervised classifiers trained on these clusters reached up to 99% accuracy (logistic regression) and 97.5% (support vector machine (SVM)), validating the discovered groupings. This work introduces a novel, data-driven methodology integrating unsupervised clustering with supervised prediction, providing a practical framework for designing immersive VR learning environments
Advances in AI, IoT, and smart systems for emerging electrical and computer engineering applications
Get PDFThe current issue of the International Journal of Electrical and Computer Engineering (IJECE) showcases a diverse array of research at the intersection of artificial intelligence (AI), Internet of Things (IoT), machine learning (ML), and advanced engineering systems. Highlighted studies explore the application of autonomous mobile robots for logistics and material handling, sensorless control and acceleration profiling of electric drives, hybrid control strategies for high-performance electric vehicles, and deep learning methods for image recognition, emotion detection, and software fault prediction. Further contributions address practical implementations of IoT in heatstroke prevention, hydroponics, Spirulina cultivation, and energy-efficient greenhouse management, demonstrating how intelligent systems can optimize resource use, safety, and productivity. The issue also emphasizes AI-empowered modeling in accelerator design, solar photovoltaic power forecasting, and GIS automation, while exploring cybersecurity through intrusion detection frameworks and fraud detection in financial systems. Cutting-edge deep learning models such as convolutional neural networks (CNN), vision transformers, and TinyML are leveraged for healthcare, nuclear monitoring, and prenatal diagnostics. Collectively, these contributions underline the transformative role of AI, IoT, and hybrid intelligent systems in electrical and computer engineering, bridging theoretical advances with practical, real-world applications. This issue aims to inspire continued research and development toward efficient, secure, and adaptive technologies that advance smart engineering solutions worldwide
IDPS: A machine learning framework for real-time intrusion detection and protection system for malicious internet activity
Get PDFWith the increasing frequency and complexity of cyber threats, there is a pressing need for effective real-time solutions to detect and prevent malicious activities. This study introduces a novel machine learning-based architecture for real-time cybersecurity to enhance accurate identification and prevention of malicious cyber activities. The proposed framework combines advanced machine learning algorithms with Wireshark network traffic analysis to effectively detect and classify a wide range of cyberattacks, providing timely and actionable insights to cybersecurity professionals. A core component of this system is a prototype blocker, which is seamlessly integrated with Cisco infrastructure, enabling proactive intervention by blocking suspicious IP addresses in real-time. In addition, a user-friendly web application enhances system operability by offering intuitive data visualization and analytical tools, enabling rapid and informed decision-making. This comprehensive approach not only strengthens network security and protects digital assets but also equips defenders with the capability to respond effectively to the dynamic landscape of cyber threats
Application of deep learning and machine learning techniques for the detection of misleading health reports
Get PDFIn the current era of vast information availability, the dissemination of misleading health information poses a considerable obstacle, jeopardizing public health and overall well-being. To tackle this challenge, experts have utilized artificial intelligence methods, especially machine learning (ML) and deep learning (DL), to create automated systems that can identify misleading health-related information. This study thoroughly investigates ML and DL techniques for detecting fraudulent health news. The analysis delves into distinct methodologies, exploring their unique approaches, metrics, and challenges. This study explores various techniques utilized in feature engineering, model architecture, and evaluation metrics within the realms of machine learning and deep learning methodologies. Additionally, we analyze the consequences of our results on enhancing the efficacy of systems designed to detect counterfeit health news and propose possible avenues for future investigation in this vital area
Parameter-efficient fine-tuning of small language models for code generation: a comparative study of Gemma, Qwen 2.5 and Llama 3.2
Get PDFLarge language models (LLMs) have demonstrated impressive capabilities in code generation; however, their high computational demands, privacy limitations, and challenges in edge deployment restrict their practical use in domain-specific applications. This study explores the effectiveness of parameter efficient fine-tuning for small language models (SLMs) with fewer than 3 billion parameters. We adopt a hybrid approach that combines low-rank adaptation (LoRA) and 4-bit quantization (QLoRA) to reduce fine-tuning costs while preserving semantic consistency. Experiments on the CodeAlpaca-20k dataset reveal that SLMs fine-tuned with this method outperform larger baseline models, including Phi-3 Mini 4K base, in ROUGE-L. Notably, applying our approach to the LLaMA 3 3B and Qwen2.5 3B models yielded performance improvements of 54% and 55%, respectively, over untuned counterparts. We evaluate models developed by major artificial intelligence (AI) providers Google (Gemma 2B), Meta (LLaMA 3 1B/3B), and Alibaba (Qwen2.5 1.5B/3B) and show that parameter-efficient fine-tuning enables them to serve as cost-effective, high-performing alternatives to larger LLMs. These findings highlight the potential of SLMs as scalable solutions for domain-specific software engineering tasks, supporting broader adoption and democratization of neural code synthesis
Systematic review of a business model using blockchain technology for the use of digital money in mass centers
Get PDFIn recent years, commercial transactions have experienced a radical change in the way goods and services are purchased. Payments with electronic and digital money are increasing dramatically compared to payments with physical money. Likewise, money using blockchain technology is marking disruptive milestones in transactions, especially in cross-border payments, showing many benefits, such as speed, lower costs, and security. The COVID-19 pandemic has shown the entire world the potential and possible development horizon of digital money, especially cryptoassets, in commercial transactions, as well as the risks associated with this technology. This has exposed the problem and need for a commercial model with blockchain technology for use in mass centers, which allows for the widespread and democratization of blockchain technology in mass commercial transactions. The methodology used is PRISMA. The objective of this article is to conduct a systematic review of the literature on digital money with blockchain technology for use in mass marketing centers. Finally, the results are presented, where the commercial model based on blockchain must consider security criteria, technology, legal aspects, and sociocultural barriers. Incorporate the interaction between electronic money, central bank digital currencies (CBDCs), and cryptoassets, as well as a decentralized technological platform for direct digital commerce. This implies that the model must consider these criteria in its design, implementation process, and the platform it supports
From YOLO V1 to YOLO V11: comparative analysis of YOLO algorithm (review)
Get PDFObject detection in images or videos faces several challenges because the detection must be accurate, efficient and fast. The you only look once (YOLO) algorithm was invented to meet these criteria. But with the creation of several versions of this algorithm (from V1 to V11), it becomes difficult for researchers to choose the best one. The main objective of this review is to present and compare the eleven versions of the yolo algorithm in order to know when using the appropriate one for the study. The methodology used for this work is aligned with preferred reporting items for systematic reviews and meta-analyses (PRISMA) principles and the results demonstrate that the choice of the best version mainly depends on the priorities of the study. If the study prioritizes accuracy and detection of small objects, it should use YOLO V4, YOLO V5, YOLO V6, YOLO V7, YOLO V8, YOLO V9, YOLO V10 or YOLO V11. While studies that prioritize detection speed should use YOLO V5, YOLO V6, YOLO V7, YOLO V8, YOLO V10 or YOLO V11. In complex environment, researchers should avoid using YOLO V1, YOLO V2, YOLO V3, YOLO V5, YOLO V7 and YOLO V9. And researchers who are looking for a good accuracy and speed and a reduced number of parameters should use YOLO V10 or YOLO V11
Image classification using two neural networks and activation functions: a case study on fish species
Get PDFLake Toba is utilized for aquaculture fishing as a clear example of how this technology can be applied. One of the species presents is the red devil fish (Amphilophus labiatus), which is known to have started appearing in the last 10 years. This species is known to be very aggressive and damage the ecosystem. When their populations go unchecked, red-devils can cause a decline in local fish populations, potentially destroying the balance of the food chain in those waters. This study used artificial neural network (ANN) and convolutional neural network (CNN) algorithms to successfully create two classification models for fish species from Lake Toba: red devil fish (Amphilophus labiatus), mujahir fish (Oreochromis mossambicus), sepat fish (Trichogaster trichopterus). The purpose of this model is to automatically identify fish species by using image-based classification techniques. According to the study's findings, both models performed exceptionally well and had a high degree of accuracy. This study addresses the lack of effective automated fish classification systems for ecosystems like Lake Toba, Indonesia, which are threatened by invasive species such as the red devil fish. By comparing CNN and ANN models with different activation functions and optimizers, we found that CNN with rectified linear unit (ReLU) activation and Adam optimizer provides the most accurate and stable results. The findings offer practical implications for fisheries management and biodiversity conservation
Accessibility in e-government portals: a systematic mapping study
Get PDFIn recent years, several researchers have investigated the challenges of accessibility in e-government portals and have contributed to many proposals for improvements. However, no comprehensive review has been conducted on this topic. This study aimed to survey and synthesize the published work on the accessibility of e-government portals for people with disabilities. We carried out a review using a systematic mapping study (SMS) to compile previous findings and provide comprehensive state-of-the-art. The SMS collected studies published between January 2000 and March 2025 were identified using an automated search in five known databases. In total, 112 primary studies were selected. The results showed a notable increase in interest and research activities related to accessibility in e-government portals. Journals are the most widely used publication channel; studies have mainly focused on evaluation research and show a commitment to inclusivity. “AChecker” and “Wave validator” are the most used accessibility evaluation tools. The findings also identified various accessibility guidelines, with the most frequently referenced being the web content accessibility guidelines (WCAG). Based on this study, several key implications emerge for researchers, and addressing them would be beneficial for researchers to advance e-government website accessibility in a meaningful way