Indonesian Journal of Electrical Engineering and Computer Science
Not a member yet
9109 research outputs found
Sort by
Advancements in seismic data collection and analysis through machine learning
Get PDFThe evolution of seismic data collection has been driven by the need for stations to capture large volumes of high-frequency signals continuously. These signals typically contain both seismic and non-seismic information. Previous research converted SEED data into CSV format and used principal component analysis (PCA) for feature extraction from the seismic dataset. Machine learning models were then employed, showing an improvement in identifying seismic and non-seismic events. This paper focuses on applying deep learning methods, specifically deep neural networks (DNN) and a hybrid model combining long short-term memory (LSTM) networks with DNN (LSTM+DNN). The proposed deep learning models demonstrate a notable improvement over traditional machine learning technique. Experimental results show a test accuracy of 99.24% using deep learning, compared to an average of 97.80% achieved with machine learning models, indicating a 1.46% enhancement in detection accuracy. This underscores the potential of deep learning in accurately detecting seismic events in real-time monitoring systems
A review and bibliometric analysis of traceability system development in the agricultural and food sector in Indonesia
Get PDFSeveral technologies and methods for traceability systems in the agriculture and food sectors have developed rapidly in recent decades. There has been an increase in traceability system research in many developing countries, including Indonesia. Our review collects data from the Scopus database to study the development and dynamics of research on traceability systems and to identify emerging technological trends in the field. This paper uses bibliometric analysis by VOSviewer to find out studies regarding traceability. Our findings reveal traceability system research in Indonesia encompasses 1,264 documents within the Scopus database from 1998 to 2022. The number of studies on traceability systems has increased significantly after 2016. Most scholarly articles on traceability technology are disseminated as conference proceedings. These traceability systems have been established and are widely adopted to ensure the quality and safety of agricultural and food products, monitor species diversity, and oversee environmental parameters. The objective of the user influences the development of the traceability system. Technologies such as deoxyribonucleic acid (DNA) barcoding, unmanned aerial vehicles (UAVs), satellites, wireless sensor networks (WSNs), blockchain, product tagging, spectroscopy, and smart packaging rapidly advance to enhance traceability capabilities
Comparative analysis of edge detection image processing techniques for efficient traffic signal management system
Get PDFTraffic monitoring and control pose significant challenges, particularly in metropolitan areas worldwide. Traditional methods such as timers, traffic police control, and sensor-based systems have become increasingly ineffective due to escalating traffic volumes. Image processing emerges as a promising technology for enhancing traffic control systems. This research aims to address the inefficiencies of current traffic management systems (TMS) by conducting a detailed comparative analysis of conventional image processing techniques, focusing on edge-based methods. Utilizing the open source computer vision (OpenCV) Library, we evaluated various edge detection techniques based on qualitative parameters like environmental lighting, object detection, and size, as well as quantitative parameters such as average traffic density, sharpness ratio (SR), abruptness, sensitivity factors (SF), processing time, frame processing time, and average frames per second (FPS). The study finds that the Canny edge detection technique outperforms others, with an average traffic density of 30.12 across 10 frames, superior SR, and minimal processing time of 99 seconds. This makes it highly effective for traffic control applications by generating high-quality edge maps with minimal noise. Our findings suggest that improving conventional image processing techniques and integrating deep learning algorithms can further enhance TMS, leading to more efficient urban mobility and reduced environmental impact
Implementation of a prototype to prevent childhood accidents in dangerous domestic environments using ESP 32 Wi-Fi module
Get PDFRobotics has significantly advanced human evolution by optimizing tasks in fields such as medicine, engineering, and mechanics, enhancing daily life through various robotic prototypes. These innovations help prevent accidents and injuries, whether at home or in hazardous environments. For instance, sensors can detect gas leaks, fires, and other potential disasters. This research aims to design a prototype adaptable to any home environment that poses risks to infants, such as kitchens, bathrooms, or stairs. The proposed prototype incorporates gas, motion, and sound sensors connected to a Wi-Fi ESP 32 module, which alerts parents to any potential danger to their children. The research is developed in six phases: component selection, circuit simulation, prototype design, three-dimensional (3D) printing, code programming, and final testing. The results demonstrate a positive impact, improving the control and care of infants by alerting parents to hazards such as gas leaks, crying, or movement in risky areas. The conclusion confirms the effectiveness of the prototype in providing timely alerts to safeguard infants in potentially dangerous situations
An ensemble image augmentation approach to enhance granular parakeratosis dataset
Get PDFThe study discusses the revolutionizing impact of deep convolutional neural network (CNN) techniques on medical image classification, particularly in identifying skin lesions. It addresses the challenge of limited datasets for granular parakeratosis (GP) and paraneoplastic pemphigus (PNP) by employing traditional and advanced ensemble data augmentation techniques. These techniques include geometric transformations, generative adversarial networks (GANs), Cutout, and keep augment. GP affects keratinization in the groin and other regions, while PNP is associated with malignancies. The study’s relevance is enhanced by the shared imaging characteristics of the chosen conditions. By utilizing tools like U-net for segmentation, region props for feature extraction, and a support vector machine (SVM) 10-fold cross-validation model for classification, the study achieved impressive performance metrics, including 95% accuracy, 100% sensitivity, and 100% specificity when evaluated on the DermnetNZ skin lesion dataset. These findings underscore the effectiveness of augmentation in enhancing the precision of medical image classifiers and signify a substantial improvement over traditional method. Thus, the research showcases the critical role of data augmentation in overcoming data scarcity challenges and advances medical image analysis
HCRF: an improved random forest algorithm based on hierarchical clustering
Get PDFRandom forest (RF) selects feature subsets randomly. Useless and redundant features will lower the quality of the selected features and subsequently affect the overall classification accuracy of the RF. This study proposes an improved RF algorithm based on hierarchical clustering (HCRF). The algorithm uses hierarchical clustering algorithms to optimize the feature selection process, by establishing similar feature groups based on the GINI index, and then selecting features from each group proportionally to construct the feature subset. The feature subset is then used to construct a single classifier. This process increases the filtering of feature subsets, reducing the negative impact of useless and redundant features on the model, and improving the model's generalization ability and overall performance. In the experimental verification, ten datasets of different sizes and domains were selected, and the accuracy, precision, recall, F1 score, and running time of HCRF, support vector machine (SVM), RF, classification and regression tree (CART) were compared using 10-fold cross-validation. Combining all the results, the HCRF algorithm showed significant improvements in all evaluation indicators, proving that its performance is superior to the other three classifiers. Therefore, this algorithm has broad application areas and value, and effectively improves the overall performance of the classifier within a lower complexity range
Application of quantum annealing solvers along with machine learning algorithms to identify online deception
Get PDFThe rising frequency of online transactions has heightened the potential of online fraud, posing significant concerns for consumers, organizations, and financial institutions. Conventional fraud detection systems frequently inadequately handle the dynamic and shifting characteristics of fraudulent activity. The increasing menace of online fraud requires novel strategies to improve the effectiveness of fraud detection systems. This study has developed and implemented a detection framework utilizing a quantum machine learning (QML) technique that integrates support vector machines (SVM) with quantum annealing solvers. We assessed its detection performance by comparing the QML application's efficacy against twelve distinct ML techniques. This study examines the integration of classical ML algorithms with quantum annealing solutions as an innovative approach to enhance online fraud detection. This study examines the possible integration of ML and quantum computing to tackle the rising issues of fraudulent activities in online transactions, as existing solutions are inadequate. This work seeks to illustrate the viability and efficacy of using these technologies, including quantum annealing to enhance the intricate decision-making processes involved in fraud detection. We offer insights on the performance, speed, and adaptability of the integrated model, highlighting its potential to transform online fraud detection and enhance cyber security measures
An ensemble approach for detection of diabetes using SVM and DT
Get PDFAs diabetes affects the health of the entire population, it is a chronic disease that is still an important worldwide health issue. Diabetes increases the possibility of long-term complications, such as kidney failure and heart disease. If this disease is discovered early, people may live longer and in better health. In order to detect and prevent particular diseases, machine learning (ML) has become essential. An ensemble approach for detection of diabetes using support vector machine (SVM) and decision tree (DT) presents in this paper. In this case, to identify diabetes, two ML techniques are DT and SVM have been combined with an ensemble classifier. They obtain the information, they require from the Public Health Institute’s statistics area. There are 270 records, or instances, in the collection. This dataset includes the following attributes: age, a body mass index (BMI) glucose, and insulin. The development of a system that predictions a patient’s risk of diabetes is the goal of this analysis. Several performance metrics, including F1-score, recall, accuracy, and precision, were used to achieve this. From overall results, 96% of precision, 97% of accuracy, 96% of F1-score, and 97% of recall values are the results achieved for the ensemble model (SVM+DT) which is more effective than other individual ML models as DT and SVM
Kimball data warehouse for the sales analysis process in a manufacturing business in Perú
Get PDFThe main goal of this research is to demonstrate that the use of innovative technology like business intelligence (BI) in a specific type of business significantly impacts their sales processes, enhancing decision-making, promotional strategies, and consequently customer loyalty and sales growth. The case study is a manufacturing business located in Lima, Peru. The information requirements of this business were analyzed, and a data mart model was created using the Kimball methodology. This multidimensional model enabled the comparison of client sales trends to propose new promotions and marketing strategies. The data analysis used to evaluate the results included hypothesis testing, analysis of employee responses to questionnaires to measure the impact of technology use on sales processes, and data reviews to assess sales increases both before and after the implementation of this technology. In both cases, the approval of the BI technology by the employees was satisfactory, and the increase in sales quantity was significant
A deep learning model with an inductive transfer learning for forgery image detection
Get PDFDue to the availability of affordable electronic devices and several advanced online and offline multimedia content editing applications, the frequency of image manipulation has increased. In addition, the manipulated images are presented as evidence in courtrooms, circulated on social media and uploaded upon authentication to deceive the situation. This study implements a deep learning (DL) framework with inductive transfer learning (ITL) by using a pre-trained network to benefit from the discovered feature maps rather than starting from scratch and fine-tuning the process to check and classify whether the suspected image is authenticated or forged effectively. To experiment with the proposed model, we used both Columbian uncompressed image splicing detection (CUISD) and the CoMoFoD dataset for training and testing. We measured the model’s performance by changing hyperparameters and confirmed the better selection of values for the hyperparameter to yield compromised results. As per the evaluation results, our model showed improved results by classifying new instances of images with an average precision of 89.00%, recall of 86.43%, F1-score of 87.32, and accuracy of 87.72% and consistently performed better compared to other methods currently in use