Indonesian Journal of Electrical Engineering and Computer Science
Not a member yet
9109 research outputs found
Sort by
Long-term user engagement in recommender systems: a review
Get PDFThe purpose of recommender systems (RS) is to facilitate user collaboration and communication on the platform. Nevertheless, there is limited knowledge regarding the extent of this relationship and the techniques by which RS could promote persistent user engagement with the platform. In order to fill this void, the present study investigates the role of RS in transforming users’ short-term engagement with the RS into long-lasting involvement with the platform. We present a theoretical framework by reviewing relevant literature in the domains of RS and user engagement to probe these issues. We provide open challenges in this field along with metrics in the present study
Feasibility study and simulation of a 4 MW solar power plant design using PVSyst in Oefafi Village, Kupang, Indonesia
Get PDFIndonesia has a vast amount of potential for renewable energy, with many places still experiencing inaccessibility and inadequacy to electricity. One of the places, Kupang in East Nusa Tenggara, suffers a deficit of 4 MW at night along with a blackout despite the implementation of the Oelpuah Solar Power Plant that can produce 5 MW. Thus, this study presents a feasibility study and simulation of additional solar power plant implementation using PVSyst based on geospatial analysis by ArcGIS to point down the desired location, namely Oefafi. The results produce 7,212,139 kWh/year (almost 20 MWh/day using 6,552 modules of Jinkosolar 610 Wp JKM-610N-78HL4- BDV, 274 units of Growatt 12KTL3-X inverters, and 2000 BYD B-Box PRO 2.5 1,024 V 5,200 Ah battery. The losses of the systems mostly come from temperature loss in PV. The total installation cost is 51 billion IDR with a payback period of 8.3 years, NPV of 6.6 billion IDR, and IRR of 10.62% for an electricity tariff of 1,000 IDR/kWh. The amount of greenhouses saved accounts for 134,110.3 tCO2. This will help in filling the 4MW deficit while also potentially providing to other unsupplied regions. Future research may use other software for comparison and give better explanations
Enhancing vocational computer engineering education with a GPT-driven speech recognition tool
Get PDFThis research investigates the effectiveness of an AI-driven speech recognition and GPT-powered learning tool in enhancing vocational students’ proficiency in computer networks. The study involved 100 students from vocational hig school, who used the prototype as part of their learning process. A pre-test/post-test design was employed to assess changes in proficiency, and students also provided feedback on the tool’s usability and impact. The results showed a consistent improvement in proficiency across all classes. A strong positive correlation was found between students’ feedback and their proficiency improvement, suggesting that students who rated the prototype as Very Helpful were more likely to see significant learning gains. However, the correlation between time spent using the tool and proficiency improvement was minimal, indicating that the quality of engagement with the tool was more important than the duration of usage. These findings highlight the prototype’s potential to improve vocational learning outcomes and underscore the importance of user satisfaction in driving success, with future refinements necessary to ensure the tool’s broader effectiveness across different learning contexts
Graphene-based reconfigurable FSS for dynamic millimeterwave OAM beam generation
Get PDFThis research paper explores the dynamic generation of orbital angular momentum (OAM) beams at millimeter-wave (mm-wave) frequencies using intelligent reconfigurable metasurfaces (IRM). The ability to dynamically control OAM properties is crucial for unlocking these beams’ full potential. This paper proposes a novel method utilizing a frequency-selective surface (FSS) integrated with reconfigurable graphene to generate an IRM. By carefully designing the FSS elements and controlling the graphene’s electrical conductivity, the system can generate and manipulate mm-wave OAM beams with different topological charges. With the suggested IRM structure, a conversion efficiency of nearly 80% can be achieved in converting the circularly polarized incident wave into its cross-polarized component at 30.7 GHz, with an overall thickness of 0.067 λ. This research has significant implications for advancing mm-wave communications by providing additional spatial dimensions for multiplexing and enhancing system capacity
Geographic information system for marine ecotourism and rural lifestyle in Prachuap Khiri Khan
Get PDFAccording to the Prachuap Khiri Khan Province tourism statistics report for 2023, there were 11,143,079 Thai and foreign tourists from January to December 2023, which increased by 1,395,195 people or 14.31 percent compared to 2022. Simultaneously, tourist attractions accumulated tourism income in 2023 totaling 44,241 million baht, marking an increase of 11,402.63 million baht or 34.72 percent from 2022. Despite this growth, tourist attractions that are popular with tourists remain centered in Hua Hin District due to a lack of publicity and insufficient information provided to tourists. Consequently, the researcher intended to develop a geographic information system (GIS) for marine ecotourism and rural lifestyles in Prachuap Khiri Khan Province to promote rural tourist attractions and distribute tourism income to the community. The system utilized the classification (precision and recall) model and was developed using ArcGIS and the web app builder ArcGIS. Findings from 8 experts in computers, information technology (IT), and GIS indicated that the overall system efficiency had an average of 4.54 and a standard deviation of 0.50. Additionally, results from the study on retrieval efficiency using the classification (precision and recall) model revealed a precision value of 0.90 and a recall value of 0.95
Context dependent bidirectional deep learning and Bayesian gaussian auto-encoder for prediction of kidney disease
Get PDFChronic kidney disease (CKD) has emerged as a significant global health issue, leading to millions of premature deaths annually. Early prediction of CKD is crucial for timely diagnosis and preventive measures. While various deep learning (DL) methods have been introduced for CKD prediction, achieving robust quantification results remains challenging. To address this, we propose the context-dependent bi-directional DL and Bayesian gaussian autoencoder (CDBDP-BGA) method for CKD prediction. This approach utilizes clinical parameters and symptoms from a structured dataset. By incorporating context dependence into the bi-directional long short-term memory (Bi-LSTM) model, CDBDP-BGA efficiently redistributes the representation of information, enhancing its modeling capabilities. Feature selection is optimized using a BGA-based algorithm, which employs the Bayesian gaussian function. The SoftMax activation function classifies CKD into five distinct stages based on estimated-glomerular filtration-rate (eGFR), considering both symptoms (texture and numerical features) and clinical parameters (age, sex, and creatinine). Simulation results using two datasets demonstrate that CDBDP-BGA outperforms conventional methods, achieving 97.4% accuracy without eGFR and 98.7% with eGFR
Non-contact breathing rate monitoring using infrared thermography and machine learning
Get PDFMonitoring vital physiological parameters such as breathing rate (BR) is crucial for assessing patient health. However, current contact-based measurement methods often cause discomfort, particularly in infants or burn patients. This study aims to develop a non-contact system for monitoring BR using infrared thermography (IRT). This approach permits to detects and tracks the nose from thermal video, extracts temperature variations into a breathing signal, and processes this signal to estimate BR. The estimated BR is then classified into three health categories (bradypnea/normal/tachypnea) using k-nearest neighbors (k-NN). To evaluate system accuracy and robustness, experiments were conducted under three conditions: (i) stationary breathing, (ii) breathing with head movements, and (iii) specific breathing patterns. Results demonstrated high consistency with contact-based photoplethysmography (PPG) measurements, achieving complement of the absolute normalized difference (CAND) index values of 94.57%, 93.71%, and 96.06% across the three conditions and mean absolute BR errors of 1.045 bpm, 1.259 bpm, and 0.607 bpm. The k-NN classifier demonstrated high performance with training, validation, and testing accuracies of 100%, 100%, and 99.2%, respectively. Sensitivity, specificity, precision, and F-measure results confirm system reliability for non-contact BR monitoring in clinical and practical settings
Enhancing uncollateralized loan risk assessment accuracy through feature selection and advanced machine learning techniques
Get PDFAccuracy in evaluating the risk of credit applications is crucial for lenders, particularly when dealing with unsecured loans. Accuracy can be enhanced by selecting suitable features for a machine learning model. To better identify high-risk borrowers, this study applies an elaborate feature selection technique. This study uses the light gradient boosting machine (LGBM) Classifier model with boosting type gradient boosting decision tree (GBDT) algorithm and n_estimator value 100 for feature selection process. This work uses advanced machine learning techniques namely stacking to improve accuracy model perform. The dataset consists of 307,506 applicants from European lenders who have applied for loans in Southeast Asia. Each applicant is described by 126 different features. Using GDBT algorithm GBDT, 30 best features were selected based on their maximum accuracy compared to another feature. By employing a stacking technique that combines the LGBM, gradient boosting (GB), and random forest (RF) models, and utilizing logistic regression (LR) as the final estimator, an accuracy of 0.99637 was reached. This study demonstrates an improved the accuracy compared to previous research. This discovery indicates that utilizing feature selection and stacking method can provide one of the most precise choices for modelling the binary class classification among the current models
Acute lymphoblastic leukemia diagnosis and subtype segmentation in blood smears using CNN and U-Net
Get PDFAcute lymphoblastic leukaemia (ALL) is a severe disease requiring invasive, expensive, and time-consuming diagnostic tests for definitive diagnosis. Initial diagnosis using blood smear pictures (BSP) is crucial but challenging due to the similar indications and symptoms of ALL, often leading to misdiagnoses. This study presents a custom approach using Convolutional Neural Networks (CNNs) to detect all cases and categorize subtypes. Utilizing publicly available databases, the study includes 3562 blood smear images from 89 patients. The innovative combination of U-Net for segmentation and various CNN architectures (U-Net, MobileNetV2, InceptionV3, ResNet50, NASNet) for feature extraction, with DenseNet201 being the most effective, forms the core of this method. The U-Net model achieved a segmentation accuracy of 98% by recognizing patterns within blood smear images. Following segmentation, CNN architectures extracted high-level features, with DenseNet201 proving the most effective in diagnostic and classification tasks. Our proposed custom CNN model achieved a test accuracy of 98%, with a training accuracy of 99.31% and validation accuracy of 97.09%. This approach enables an accurate distinction between ALL and non-pathologic cases
Comparative evaluation of PVGIS, PVsyst, and SAM models for predicting solar power output in equatorial tropical climates
Get PDFAccurate evaluation of energy production in photovoltaic (PV) systems is critical for renewable projects, especially in tropical climates where environmental factors such as temperature significantly affect performance. Although commercial simulation tools exist (photovoltaic geographic information system (PVGIS), PVsyst, and system advisor model (SAM)), previous studies have identified notable deviations between their predictions and actual data, particularly in tropical climates. Moreover, these investigations are usually limited to short periods (one year) and do not systematically compare multiple tools under interannual conditions. This study evaluates the accuracy of PVGIS, PVsyst, and SAM in predicting the energy production of a PV installation in a tropical equatorial climate for 24 months to identify the most suitable tool for this context. Monthly energy production data were collected from a PV plant in Monteria, Colombia, equipped with 240 modules and two 36 kW inverters. Simulations were performed using the most recent PVGIS, PVsyst, and SAM versions. Accuracy was evaluated using metrics such as root mean square error (RMSE) and mean absolute error (MAE). SAM showed the highest accuracy, with an overall RMSE of 1,993.71 kWh and MAE of 1,615.87 kWh, followed by PVGIS (RMSE: 2,076.65 kWh, MAE: 1,830.84 kWh) and PVsyst (RMSE: 3,546.18 kWh, MAE: 3,250.17 kWh). The results highlight that SAM provides estimates closer to the real data and less dispersion than other tools. This study contributes to the renewable energy field by systematically comparing simulation tools in an understudied tropical context. The findings emphasize the importance of selecting appropriate software according to the specific environmental conditions of the project, thus optimizing the design and efficiency of PV systems in tropical regions