Indonesian Journal of Electrical Engineering and Computer Science
Not a member yet
9109 research outputs found
Sort by
Ensemble recursive feature elimination-based ensemble classification for medical diagnosis
Get PDFThe application of data mining techniques for the extraction of patterns from medical datasets is useful in the prediction of various diseases from the data of patients. An appropriate feature selection method is required for the medical datasets to give better results for the medical data mining process. In data preprocessing, feature selection is an important process that finds the most relevant features from the dataset. Considering all features of the medical dataset without using any feature selection process may sometimes lead to inaccurate results. Most of the medical datasets contain meaningless data that are not relevant to the data mining process. These data can be eliminated through the feature selection process. This paper presents an integration of an ensemble feature selection approach and an ensemble classification approach through a classifier called the ensemble recursive feature elimination-based ensemble classifier (ERFE-EC) for the classification of medical data. Four different medical datasets were used for testing the ERFE-EC method, which showed promising results
Improved YOLOv8 for rail squat detection and identification
Get PDFRail transport plays a vital part in the country's economy by ensuring the safe movement of both goods and passengers. Therefore, maintaining rail safety through consistent surface defect inspection is extremely importan. However, squat defect detection on rail surfaces faces considerable difficulties due to weather impacts, lighting changes, and variations in image contrast. These challenges hinder the accuracy and reliability of traditional inspection methods. To solve this problem, this study proposes an improved YOLOv8 model for the identification and classification of squat defects. The methodology involves capturing images of the rail track, preprocessing them to enhance image quality, labeling squat defects for training purposes, and training the proposed model using the labeled dataset. The improved YOLOv8 model incorporates enhancements such as multi-scale convolution modules and attention mechanisms to improve feature extraction and defect recognition. Experimental results demonstrate the effectiveness of the proposed method, achieving an impressive accuracy of 0.92 in detecting and categorizing squat defects. These findings highlight the potential of the proposed approach to enhance railway safety by providing a reliable and efficient solution for rail surface inspection
Generation of distribution routes with shorter distances and fewer vehicles using the simulated annealing algorithm
Get PDFThe vehicle routing problem (VRP) is still a persistent challenge in society, and can be considered a combinatorial optimization problem, where a fleet of delivery vehicles must satisfy the demand of customers sharing the same depot, minimizing the transport distance. The objective of this research is to propose a method to generate distribution routes that minimize both the number of vehicles used and the total distance traveled. To this end, an initial solution is used, on which the Greedy algorithm is applied, followed by the simulated annealing (SA) algorithm, manipulating the exchange techniques, insertion methods, parameter adjustments within the algorithm and applying the penalty as a mechanism to avoid the excessive use of trucks or the assignment of routes that exceed the allowed capacity. The proposal was validated using four datasets, as a result, the general averages of the reduction in distance, changes and penalty cost are shown: The Greedy algorithm reduced the distance by 5.71%, in trucks to 16.57%, in penalty cost to 14.71%; then, applying the SA algorithm, a better efficiency was achieved by reducing the distance by 10.36%, 20.08% in trucks and 18.64% in penalty cost. In this way, the use of vehicles in the distribution routes is optimized, which could contribute to the reduction of vehicular traffic and the reduction of CO2 emissions, thus favoring the environment
Development of an analysis capacity model for high electron mobility transistor AlGaN/GaN
Get PDFIn this paper, we demonstrate the analytical model developed to characterize the gate-to-drain capacitance Cgd and the gate-to-source capacitance Cgs, and the impact of the gate length on those capacitances, for the high electronic mobility transistor based on GaN. This model is developed from our previous work on the current voltage characteristic (I, V), and small signal parameters for AlGaN/GaN HEMT. The research study examined the impact of parasitic resistances (drain, source), low field mobility, the aluminum amount in the AlGaN barrier, and high-speed saturation. The developed model has matched the experimental data well, confirming the validity, accuracy, and robustness of the model we have developed
Enhancing document text classification using hybrid deep contextual and correlation network
Get PDFDocument analysis involves the extraction and processing of information from documents, a task increasingly automated through the use of deep learning (DL) technologies. Despite the high predictive power of DL models, their black-box nature poses challenges to transparency and interpretability, hindering their integration into the industry. This paper introduces the hybrid deep contextual and correlation network (HDCCNet), a novel methodology designed to improve both the accuracy and interpretability of multi-category classification tasks. HDCCNet leverages a hybrid layer category correlation module to deepen category connections, thereby enhancing the understanding and prediction of category interrelations. To address potential prediction divergence, residual connections are incorporated, ensuring stable and reliable performance. Furthermore, HDCCNet reduces model parameters, accelerating convergence and making the model more efficient. This efficiency is particularly beneficial for practical applications, allowing faster deployment and scalability. By bridging the gap between DL’s capabilities and industry needs for transparency, HDCCNet provides a robust solution for automated document processing, paving the way for broader adoption of DL technologies in business environments
Mobile application for diagnosing alzheimer's based on clinical dementia rating
Get PDFAlzheimer's is a neurodegenerative disease characterized by memory loss, impaired thinking abilities, and changes in behavior. It is the most common form of dementia, significantly affecting a person's ability to carry out daily activities. Statistics indicate that the number of individuals suffering from Alzheimer's worldwide continues to rise as the population ages. Diagnosing Alzheimer's is a complex process that typically requires a skilled medical team. One diagnostic tool that can be utilized is an MRI machine. Previous research focused on extracting features from MRI images taken from three different cross-sections: axial, coronal, and sagittal. Based on these three types of cross-sectional images, we developed a system to classify the severity of Alzheimer's. This paper focuses on creating an Alzheimer's classification system accessible through a mobile application. The results indicate that our system has a performance accuracy of 90% in classifying the severity of the disease
Proposition of a new fitness function: Hadj-said fitness function
Get PDFIn the dynamic field of artificial intelligence, genetic algorithms (GAs) offer a powerful approach to solving complex problems by mimicking biological mechanisms such as mutation, crossover, and natural selection. Their efficiency relies primarily on the fitness function, which evaluates the quality of candidate solutions and guides the evolutionary process toward an optimal outcome. A well-designed fitness function not only enhances convergence speed but also reduces the risk of stagnation and improves algorithmic accuracy. This paper explores the fundamental role of fitness functions in optimization, machine learning, multi-objective optimization, and cryptography, highlighting their impact on the performance of GAs. We propose a novel fitness function that incorporates the influence of crossover, mutation, and inversion rates on solution quality. This approach, which diverges from conventional models, demonstrates improved convergence behavior and adaptability across different problem domains. The proposed method enhances GA performance not only in secure data encryption but also in general optimization and learning tasks, making it a valuable contribution for both researchers and practitioners, which can open new avenues for research in the development of more robust evolutionary strategies that can adapt effectively to the specific characteristics and challenges of each problem domain
Low complexity blind selective mapping in orthogonal frequency division multiplexing: utilizing linear combination
Get PDFOrthogonal frequency division multiplexing (OFDM) is a cornerstone in wireless communications for its spectral efficiency and robustness against multipath fading. However, its deployment is constrained by the high peakto-average power ratio (PAPR), which demands complex power amplifiers and increases system costs. Selective mapping (SLM) is a popular distortion less method for PAPR reduction but suffers from high computational complexity and data rate losses due to side information (SI) transmission. This paper proposes a low-complexity, blind SLM method utilizing linear combination, which reduces computational complexity by generating alternative candidate signals without additional inverse fast fourier transform (IFFT) operations. A maximum likelihood estimation (MLE)-based blind receiver recovers transmitted signals without SI, preserving data rate integrity. The proposed method achieves comparable PAPR and bit error rate (BER) performance to conventional SLM (C-SLM) while significantly reducing computational operations. Simulations demonstrate the efficiency of the method across various configurations, making it a strong candidate for next-generation communication systems like 5G and beyond
Personal identification system based on multidimensional electroencephalographic signals
Get PDFPersonal authentication using electroencephalographic (EEG) signals, is one of the important applications in brain computer interface (BCI). In this work we investigate the use of EEG signals as a biometric trait. Multidimensional EEG signals were represented as symmetric positive-definite (SPD) matrices on a Riemannian manifold. Two experiments are performed in the first; we use minimum distance to Riemannian mean (MDRM) as a classifier. In the second; SPD matrices are vectorized, and the generated vectors are used to train various machine learning (ML) classifiers. MDRM classifier achieved a correct recognition rate (CRR) of 96.92% , while ML classifiers achieved CRR from 95.39% to 99.45%
Inset-fed microstrip patch antenna optimization for 2.4 GHz using surrogate model assisted differential evolution machine learning algorithm
Get PDFIn this work, we have used the surrogate model assisted differential evolution (SADEA) to model a one and two-element inset-fed patch antenna array to optimize its parameters for efficiency and usability. The microstrip patch antennas operates in a frequency band of 2.4 GHz. The optimization process focused on fine-tuning the patch length, patch width, and notch width to enhance key performance metrics directivity, return loss, and bandwidth. The design is made in CST software with an FR-4 substrate and simulated in the ADE1.0 software a MATLAB toolbox. Significant enhancements were achieved including a directivity gain of 3.04 dB, and 5.58 dB a return loss of -19 dB, -16 dB, and an expanded impedance bandwidth from 0.0798 GHz, 0.0588 GHz to 0.0951 GHz, 0.0824 GHz respectively. The antenna was constructed and then measured. The findings showed that the measurements and the fabrication process closely matched, especially in terms of return loss