Bulletin of Electrical Engineering and Informatics
Not a member yet
2885 research outputs found
Sort by
Creating and analysing privacy policies of Malaysia e-commerce using personal data protection act
Full text linkDespite legally binding agreements between users and website owners, users often overlook website privacy policies due to their length and complexity. Transparency in these policies is crucial, particularly in Malaysia, where regulatory agencies face challenges ensuring compliance with the personal data protection act (PDPA) of 2010 due to intricate language and complex legal clauses. Machine learning has been used to analyse privacy policies under various legal frameworks, but no dataset currently exists for the Malaysian PDPA. Thus, to bridge this gap, we introduce a pilot corpus of 50 privacy policies specifically tailored to the Malaysian PDPA. This dataset is analysed and made available for academic research, offering insights into privacy regulations and identifying trends in privacy policy transparency. Our findings pave the way for the development of tools to enhance compliance with PDPA standards and improve policy readability for users. The corpus also serves as a foundation for further research in privacy and data protection, encouraging the exploration of automated approaches for policy analysis and regulatory oversight
Improved non-invasive diagnosis of hepatocellular carcinoma by optimized meta classifier with hybridized features
Full text linkHepatocellular carcinoma (HCC), the primary cancer of the liver, is life-threatening, with few or no symptoms, and detection in the early stage will help for successful treatment with surgery, and transplant for a better life quality. Here, we proposed two stacking classification models based on deep learning with differential hybrid feature selection for the early detection of HCC using novel non-invasive biomarker PIVKA-II. We showed how the variations in hybrid feature selection affect the performance of stacking classification and different supervised machine-learning algorithms on a metaclassifier. The base layers were support vector machine (SVM), gradient boosting (GB), and linear discriminant analysis (LDA). The meta classifier was a multilayer perceptron (MLP) with three different optimizers, stochastic gradient descent (SGD), adaptive moment estimation (ADAM), and Limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS). Our first model outperformed the second with their hybrid features by improving accuracy by 1.5% and F1_score by 1% in both SGD and ADAM optimization, while MLP-LBFGS had a 1.4% increase in accuracy. The precision had hiked by 1.9%, 3.5%, and 1.7% in SGD, ADAM, and LBFGS, respectively, in Model-1. Matthew’s correlation coefficient (MCC) for MLP-SGD increased from 0.82 to 0.85, MLP-ADAM from 0.81 to 0.85, and MLP-LBFGS from 0.75 to 78 for the first model
An efficient inductor-capacitor-inductor-capacitor compensation topology for wireless power transfer system
Full text linkWireless power transfer (WPT) systems provide a promising alternative for charging various applications, including electric vehicles (EVs), biomedical implants, smartphones, and network sensors. However, these systems often struggle to maintain high efficiency under varying loading and coupling conditions. This paper addresses these challenges by proposing a novel hybrid inductor-capacitor-inductor-capacitor (LC-LC) compensation topology. The proposed LC-LC topology is specifically designed to outperform conventional single-element compensation topologies, such as series-series (SS) and series-parallel (SP) configurations, by effectively reducing leakage inductance between coils. An analytical model of the LC-LC topology is developed and validated through simulations using Keysight advanced design system (ADS) software. The results demonstrate that the LC-LC topology not only achieves a peak efficiency of 99.6% under optimal conditions but also maintains superior performance compared to SS and SP topologies, with only a slight decrease to 93% efficiency observed at low load resistances. These findings highlight the potential of the LC-LC topology to significantly enhance WPT system efficiency across a range of operating conditions
Development of water quality monitoring system for fish farming
Full text linkTilapia fish farming faces growing challenges from climate variability, environmental degradation, and the urgent demand for sustainable food production. However, traditional water quality monitoring methods remain manual and reactive, often resulting in compromised fish health and reduced farm productivity. Addressing this need, this study designed and developed a water quality monitoring system utilizing the internet of things (IoT) and embedded systems to enable real-time, proactive management. Guided by the software development life cycle (SDLC), the methodology focused on planning and analysis, system design and development, and testing and evaluation. The system integrates key water quality sensors, including pH, temperature, dissolved oxygen (DO), and electrical conductivity (EC), identified as critical parameters affecting tilapia health. These sensors were interfaced with Arduino Nano and ESP32 Dev Kit microcontrollers, forming the sensing layer of the system. Sensor data were transmitted to the ThingSpeak IoT platform for real-time visualization and storage. Validation results revealed a low mean absolute percentage error (MAPE), indicating an acceptable sensor performance. User evaluation, based on the technology acceptance model (TAM), indicated that the system was perceived as useful, user-friendly, and valuable for aquaculture management. Overall, the system enables real-time water quality monitoring, supporting a more responsive and sustainable environment for tilapia fish farming
Influences of impulse generators on the impulse characteristics of grounding systems
Full text linkIt is important to ensure the effectiveness of the experimental test set up and to accurately characterize grounding systems under high impulse conditions, the study on the effect of impulse generator is therefore needed. As with other experimental work, the test results may be influenced not only by the characteristics of the test load under study, but also the test arrangement, rating of the impulse generator and transducers. In this work, sources of this overshoot/spike observed in voltage and current traces of 1-rod, 3-rod, and 4-rod electrodes subjected to two impulse current generators of different rating: generating at maximum voltage and current of 100 kV, 1.5 kA, and 300 kV, 10 kA with the same response time of 1.2/50 μs are identified with the aid of simulation work
The effect of feature selection with optimization on taxi fare prediction
Full text linkFeature selection plays a key influence in machine learning (ML); the main objective of feature selection is to eliminate irrelevant and redundant variables in different classification problems to improve the performance of the learning algorithms. Classification accuracy is improved by reducing the number of selected features. Many real-world problems, such as taxi fare can be predicted by ML. This paper proposes feature selection using genetic algorithm (GA) optimization to predict taxi fare. Experiments are performed on real datasets of taxi fare, and this paper uses eight classifiers to evaluate the selected features. The performance of the classifiers is assessed using various performance metrics. The results are compared with feature selection without optimization. The proposed method records high classification accuracy when evaluated by three types of classifiers (random forest, AdaBoost, and Gradient Boost). The results indicate that the prediction accuracy of the proposed method is 99.7% on taxi fare dataset
Modification of grey-level co-occurrence matrix for epileptic electroencephalogram signal classification
Full text linkTexture analysis is a fundamental approach in image processing for identifying specific patterns or structures. One widely used method is the grey-level co-occurrence matrix (GLCM), which computes the frequency of pixel value pairs at certain distances and angles. This study adapts the GLCM method for 1D electroencephalogram (EEG) signal analysis, focusing on extracting features such as contrast, energy, homogeneity, correlation, and entropy. EEG signals are normalized to the range 0–255, and the extracted features are classified using a support vector machine (SVM). Experimental results show that combining features across multiple distances (d=1 to 20) achieves classification accuracy of 78.8% for five classes (Z/O/N/F/S), 94.0% for three classes (O/F/S), and 94.3% for another three-class group (Z/N/F). The method shows strong performance for short to medium distances and fewer class combinations. However, performance declines when dealing with more complex class sets and longer distances, where texture features become less effective. The drop in accuracy for Z/O/N/F/S beyond d=5 underscores the challenges of maintaining feature robustness at extended distances. Despite this, GLCM remains a promising approach for EEG signal classification. Future work should focus on optimizing distance parameters and feature combinations to further enhance classification performance
Multilayer crypto method using playing cards shuffling operation
Full text linkAn efficient and highly secure method of secret message cryptography will be presented which based on the principle of playing cards shuffling. The method will be implemented in a selected number of layers, each layer will encrypt-decrypt the input message using its own private key (PK), the output of any layer can be taken as a final encrypted-decrypted message, increasing the number of layers will increase the security level of the massage, making the hacking attacks impossible. In the encryption function a key generation and a message blocks shuffling will be executed, while in the decryption function the key generation and the message blocks reverse shuffling will be executed. The PK used in this method will be complicated and it will contain for each layer 2 chaotic parameters (r and x) and the block size (BS), utilizing these parameters, a chaotic logistic map model is run to produce a chaotic key, which is sorted to produce the layer's index key. Applying 4 layers the length of confidential key will be 768 bits, this length will be able to generate a large key space which is robust to hacking attempts. The speed parameters and throughput of the proposed will be calculated and compared with those of other methods
A novel technical analysis and survey on disaster robots for flood search and rescue operations
Full text linkThe advance in human-robot interaction brings out novel applications of the disaster rescue operations. Especially, the concept of search and rescue (SAR) assisted robot operations plays an extensive role in the natural hazards, such as earthquake and wild fires. Particularly, the SAR operations in the water-based drowning due to floods and boat capsize disaster are expensive and not fast. This paper presents a survey on various SAR based remotely operated vehicle’s (ROV) related to airborne, under and surface of the water, such as unmanned marine vehicles (UMV) and unmanned aerial vehicles (UAV). In addition, the performance analysis of each UMV such as EyeROV TUNA, Saif Seas, iBubble, DTG3, Trident, Fathom One and SEAOTTER-2, is listed which helps to select the right UMV for the rescue operation at different water depths. Also discussed various SAR based UAVs like DJI Phantom-MAVIC 2, YUNEEC-H520 Hexacopter, Microdrones MD4-1000, DSLRProsMatrice 210 RTK V2 and AltiGator’sXena Drone for the flood and boat capsize operations. However, the usage of Syma X8 Pro UAV for the flood operations are worthy than Sea King SAR Chopper, which is a cost-effective operation
Enhancing photovoltaic parameters based on modified puma optimizer
Full text linkThis article presents a photovoltaic (PV) optimization approach using the puma optimizer (PO) approach, which has been enhanced by utilizing Lévy flight optimization. The name of this approach is modified puma optimizer (MPO). PV generation systems are essential for sustainable solar energy utilization. It is an innovation and clean energy. There is an urgent demand for suitable and reliable simulation and optimization techniques for PV systems. This will result in increased efficiency. Algorithms with a high degree of reliability are needed to ensure optimal PV parameters. This study was conducted with MATLAB software. This article introduces the original PO method as a means to evaluate the performance of the MPO approach. The root mean square error (RMSE) function serves as a benchmark. Based on the simulation findings, the MPO approach shows superior RMSE compared to the PO method, specifically at a value of 0.0026%