Articles

Access the latest knowledge in applied science, electrical engineering, computer science and information technology, education, and health.

Advanced Search
Filter Icon

Filters article

Years

FAQ Arrow
Single Year
Range Year
0
0

Source Title

FAQ Arrow

Authors

FAQ Arrow

31,377 Article Results

COVID-19 infection detection using convolutional self-attention network with voting classifier

10.11591/ijaas.v15.i3.pp1072-1084
Mustafa Hussein Zwayyer , Ammar A. Ali , Rusul Hussein Hasan
Early and accurate detection of COVID-19 from chest computed tomography (CT) scans are becoming essential for effective clinical decision-making and disease control. This study is proposing a robust deep learning framework that integrates a convolutional self-attention network (CSAN), gamma correction for image enhancement, and a voting-based ensemble classifier to improving diagnostic performance. The model is being evaluated on a dataset of 2,271 CT images and is achieving an accuracy of 95.12%, sensitivity of 97.25%, specificity of 98.11%, F1-score of 96.46%, and area under the curve (AUC) of 0.977. Experimental results are demonstrating that the proposed method significantly surpasses baseline models, including standalone CSAN, residual networks 50 (ResNet-50), densely connected convolutional network 121 (DenseNet-121), and visual geometry group (VGG) 16, with improvements of up to 9.5% in accuracy. The integration of attention mechanisms, image enhancement, and ensemble learning are proving effective in capturing both local and global features, leading to more reliable classification. These findings are suggesting that the proposed framework is supporting automated COVID-19 diagnosis in real-world clinical applications.
Volume: 15
Issue: 3
Page: 1072-1084
Publish at: 2026-09-01

Recent advances in computer vision segmentation for urban tree detection using street view imagery: a literature review

10.11591/ijaas.v15.i3.pp975-989
Muhammad Fareez Rashdan , Nurbaity Sabri , Shafaf Ibrahim , Nor Masri Sahri
Segmentation of trees in urban settings is important for tracking vegetation and managing green infrastructure in urban areas. However, there are numerous factors that make it difficult to obtain precise results, such as differences in tree species, season, and environmental variables. Street-view imagery can provide a valuable database to be used for the detection and identification of trees. Unfortunately, the process of labelling images manually is time-consuming and inconsistent. In recent years, deep learning models have provided better automation and precision. However, difficulties in obtaining results persist due to factors such as occlusion, background, and changes in lighting conditions. Therefore, this paper presents an overview of the current trends in the use of deep learning for the detection and segmentation of trees. Specifically, the following aspects will be discussed in this work: preprocessing of images, semantic and instance segmentation, models based on deep learning, and transformers. This literature review also analyzes current datasets, evaluation measures, and applications, offering knowledge about existing challenges and future directions for further studies. The article is designed to be a source of reference for scientists and practitioners working on problems related to urban vegetation classification and computer vision.
Volume: 15
Issue: 3
Page: 975-989
Publish at: 2026-09-01

A novel combined deep learning-based technique for electricity theft detection in a smart grid

10.11591/ijaas.v15.i3.pp1098-1111
Etienne Mouckomey , Jacques Bikai , Camille Franklin Mbey , Felix Gislain Yem Souhe , Alexandre Teplaira Boum , Vinny Junior Foba Kakeu
Energy theft represents a major challenge for electricity distribution companies worldwide and constitutes a key component of non-technical losses (NTLs), alongside faulty meters and billing inaccuracies. Despite the rise of artificial intelligence techniques for electricity theft detection, existing approaches exhibit limitations in classification and data analysis when applied to large-scale datasets. These shortcomings can reduce the effectiveness and precision of the theft detection systems. Therefore, this article proposes a bidirectional long short-term memory (Bi-LSTM) with the lookahead adaptive moment estimation (LADAM) optimizer and a support vector machine (SVM) with a bidirectional gated recurrent unit (Bi-GRU) for the detection of energy theft in a smart electricity grid to assess energy supplier companies. In this work, the proposed model is implemented on the State Grid Corporation of China (SGCC) dataset. In the proposed framework, the Bi-LSTM is used for feature extraction, while the LADAM optimizer is employed for feature selection and normalization. The refined feature set is subsequently fed into the SVM classifier to distinguish between theft and non-theft instances. The obtained findings highlight the superior efficiency of the proposed strategy compared with those in the literature, highlighting its potential for practical deployment in real-world smart grid systems.
Volume: 15
Issue: 3
Page: 1098-1111
Publish at: 2026-09-01

Enhancing fraction multiplication and division performance through intelligent tutoring system with visual models

10.11591/ijaas.v15.i3.pp1214-1226
Ida Bagus Indrabudhi Kusuma , Selly Meliana , Dade Nurjanah
Mathematical proficiency, especially fractions, help humans to succeed in their future. However, students tend to struggle in understanding fractions, one of them is natural number bias (NNB). This study aims to address this challenge by developing a web-based math intelligent tutoring system (ITS) using the cognitive tutor authoring tools (CTATs) to create a cognitive tutor with adaptive feedback and visual models to support students in learning fraction multiplication and division. This study compares the students’ learning performance from two schools which the first school has aligned visual method with the system proposed. The results show that the experimental groups in the first school outperformed the control group with an average learning gain of 33.78%, while in the second school the control group outperformed the experimental group with average learning gain of 65.79%. Each school suggests statistically significant differences in students’ results. This highlights the effectiveness of ITS depends not only on system features but also the importance of alignment between students’ prior visual models and the visualization method in the system. Students with aligned visual method suggest positive feedback from the survey conducted, indicating its potential for improving fraction learning when it is integrated consistently with existing visual instruction.
Volume: 15
Issue: 3
Page: 1214-1226
Publish at: 2026-09-01

Drivers of salesperson performance: value transfer, personality congruence, and selling ambiance

10.11591/ijaas.v15.i3.pp1155-1165
Evi Dian Kartikasari , Sulastri Sulastri , Isnurhadi Isnurhadi , Muchsin Saggaf Sihab
This study examines preference-based transfer of value (PTV) as a mediating mechanism linking adaptive selling (AS), salesperson-brand personality congruence (SPC), and positive selling ambiance (PSA) to salesperson performance (SP) in a retail banking context. Using a quantitative cross-sectional survey, questionnaires were collected from 262 frontline salespeople of PT Bank BNI (Persero) Tbk in Indonesia. The proposed model was tested using structural equation modeling (SEM). The findings show that AS, SPC, and PSA significantly enhance PTV, and PTV positively predicts SP. Mediation analysis indicates that PTV transmits the effects of these antecedents on performance, highlighting the importance of converting selling behaviors and interaction climates into customer-internalized value. The study contributes by introducing and empirically validating PTV as a novel psychological channel in sales performance research and offers managerial implications for recruitment, training, and performance assessment in business-to-consumer financial services. Future research may extend this model across industries and examine boundary conditions using multi-group designs.
Volume: 15
Issue: 3
Page: 1155-1165
Publish at: 2026-09-01

Comparative review on control strategies for doubly fed induction generator-based wind energy harnessing system

10.11591/ijaas.v15.i3.pp1166-1192
Preethicaa Ramesh , Sowmmiya Uthayakumar , Tole Sutikno , Vigna Kumaran Ramachandaramurthy , Mochammad Facta , Auzani Jidin
Doubly fed induction generator (DFIG)-based variable-speed wind energy harvesting systems (WEHS) are widely employed owing to high efficiency and reliable grid integration under variable wind conditions. This paper presents a comprehensive review of conventional, intelligent, and hybrid control strategies for DFIG-based WEHS. Unlike previous studies that primarily focus on specific control techniques, this study systematically compares different approaches based on key performance indicators-dynamic response, total harmonic distortion (THD), robustness, and fault ride-through (FRT) capability. Conventional controllers such as proportional–integral (PI) control, field-oriented control (FOC), and direct torque control (DTC) are evaluated alongside intelligent techniques, including fuzzy logic control (FLC), artificial neural networks (ANNs), adaptive control, and optimization techniques. The comparison shows that intelligent and hybrid control strategies achieve superior dynamic response, lower THD, enhanced robustness, and improved FRT capability, although they require higher computational resources. Furthermore, this review highlights critical research challenges, including the lack of experimental validation, the absence of standardized benchmarking frameworks, and the complexity of real-time implementation in high-power WEHS. Finally, future research should focus on developing hybrid intelligent controllers, establishing evaluation methodologies, and integrating advanced control strategies with smart grid technologies to improve power quality, voltage regulation, and reliability of modern wind energy systems.
Volume: 15
Issue: 3
Page: 1166-1192
Publish at: 2026-09-01

Low-switch count based excitation of induction generator in wind power generation

10.11591/ijaas.v15.i3.pp1050-1058
Nijam Uddin Molla , Shilpi Bhattacharya , Debashis Chatterjee , Arabinda Das
The power sector is undergoing sea changes from conventional forms of power generation towards renewable energy due to diminishing resources of conventional sources of energy and also to deal with environmental issues. Three-phase self-excited induction generators (IGs) are popular in power generation from renewable sources in grid-connected as well as standalone systems. The single-phase induction generators (SPIG) with self-excitation capacitors are used in domestic applications. This paper proposes a novel reduced switch-based converter excitation system for an asymmetrical split-phase induction generators (ASPIG) application in a standalone wind energy-based power generation system. Instead of the normal six-switch inverter, the excitation scheme proposed here comprises of four numbers of power switches and four capacitors. Proper control shows that variable excitation can be effectively provided to the ASPIG using only four switches with reasonably good performance on sufficient loading and wide speed range. The proposed system is implemented and tested on load and at varied speeds in MATLAB/Simulink. This paper presents a novel excitation scheme, modeling, control, simulation, and analysis. The results obtained are reported to be effective and satisfactory in wide speed range in the standalone system.
Volume: 15
Issue: 3
Page: 1050-1058
Publish at: 2026-09-01

Amaranthus dubius seeds oil extraction using green solvent approach with terpenes

10.11591/ijaas.v15.i3.pp964-974
Rachma Tia Evitasari , Gita Indah Budiarti , Endah Sulistiawati , Adi Permadi , Ika Dyah Kumalasari , Nazira Mahmud , Nor Adila Mhd Omar
This research will explore the potential for utilizing wild amaranth (Amaranthus dubius) seeds, which are abundant and have not been utilized in Indonesia. This research will focus on optimizing the amaranth seed oil extraction process using the response surface method and the characteristics of the oil produced from amaranth seeds. The solvent used in this research is a biobased green solvent, terpenes. This research aims to explore the potential of utilizing wild amaranth seeds by optimizing the oil extraction process using a biobased green solvent, terpenes, through the response surface methodology (RSM). Fourier transform infrared spectroscopy (FTIR) results indicate that the terpenes solvent contains various terpenoids such as α-pinene, β-pinene, and camphene. By applying ultrasound-assisted extraction (UAE) and optimizing the critical process parameters (liquid-to-solid ratio, extraction temperature, and extraction time) through a Box-Behnken RSM design, the maximum oil yield was achieved under environmentally friendly conditions. The optimum condition was reached at a solvent-to-solid ratio of 5.13 ml/0.1 g, a temperature of 49.39 °C, and an extraction time of 47 minutes. With these optimal parameters, the predicted oil yield was 2.04%. The optimized extraction not only delivered a high-quality oil-rich extract but also minimized solvent residue, energy consumption, and hazardous emissions.
Volume: 15
Issue: 3
Page: 964-974
Publish at: 2026-09-01

Lesion-semantic token guided transformer fusion for stage-adaptive diabetic retinopathy grading and screening

10.11591/ijaas.v15.i3.pp1112-1122
Murali Gujjula , Kumar Narayanan
Diabetic retinopathy (DR) screening from retinal fundus images is still difficult because of the heterogeneity of lesions, significant class imbalance, and low reliability of existing deep learning algorithms owing to their generic feature pooling and passive fusion mechanisms. To overcome such limitations, this study presents a hybrid model called HandEffiTrans-DR, which combines deep retinal representations with handcrafted lesion descriptors via the directed transformer-based attention mechanism. First, a lesion-prior-guided token squeezer (LPG-TS) module is proposed in the work, which is capable of creating lesion-specific visual tokens via integrating prior knowledge about pathology into the tokenization procedure. Moreover, a directed transformer-based fusion (T→V) strategy allows for actively fusing the handcrafted lesion features with visual representations. The introduced framework is assessed with a stratified five-fold cross-validation protocol. Its average classification accuracy surpasses 97%, while the macro-F1 score is higher than 0.97 and Cohen's kappa is greater than 0.95, indicating good agreement despite the presence of class imbalance. The probability reliability is measured in terms of expected calibration error (ECE) and Brier scores and shows well-calibrated confidence. Further validation is performed on two external datasets, where consistent performance with minor decline is shown.
Volume: 15
Issue: 3
Page: 1112-1122
Publish at: 2026-09-01

Enhanced switching converter for photovoltaic systems: performance analysis and simulation

10.11591/ijaas.v15.i3.pp1262-1269
Nikunj N. Patel , Rakeshkumar A. Patel
This paper focuses on reducing switching losses and stresses in photovoltaic (PV) inverters by introducing a new five-level hybrid T-type multilevel inverter (T-MLI). The proposed topology was designed and simulated using physical security information management (PSIM) software and validated using one year of solar radiation and temperature data to account for seasonal variations. The proposed inverter was also experimentally implemented through hardware development, and its performance was evaluated under different load conditions. Under nonlinear load conditions, the results demonstrate that the proposed T-MLI reduces total harmonic distortion (THD) to 3.8%, reduces voltage stress on the switches, improves direct current (DC)-link utilization, and enhances the overall efficiency of the inverter system compared with the traditional topology.
Volume: 15
Issue: 3
Page: 1262-1269
Publish at: 2026-09-01

The role of online public relations and social influence on customer behavior at Vietnamese commercial banks

10.11591/ijaas.v15.i3.pp1015-1028
Thao Phuong Nguyen , Tuan Minh Nguyen , Hang Thi Thanh Hoang , Huong Thi Thu Nguyen
This study examines the effects of personalization (Per), social media (SM), and timeliness (Tim) on online public relations (OPR) and customer behavior (CB), as well as the mediating role of OPR and the moderating role of social influence (Soc). Data from 337 customers of Vietnamese joint-stock commercial banks were analyzed using partial least squares structural equation modeling (PLS-SEM). The results show that Per and SM significantly enhance OPR, whereas Tim does not affect OPR but directly influences CB. The indirect effects of Per, SM, and Tim on CB through OPR are not significant, indicating that OPR does not mediate the relationships. Although OPR has a small positive effect on CB in the moderation model, Soc neither directly affects CB nor moderates the OPR–CB relationship. These findings suggest that OPR supports digital relationship management but does not automatically translate communication activities into behavioral responses. Banks should therefore prioritize personalized, credible, transparent, and secure communication.
Volume: 15
Issue: 3
Page: 1015-1028
Publish at: 2026-09-01

Students' perspectives on the effectiveness of ChatGPT in virtual learning

10.11591/ijaas.v15.i3.pp921-931
Isaac Asampana , Henry Akwetey Matey , Ben T. Ocra , Jones Yeboah Nyame
This research investigates how students perceive chat generative pre-trained transformer (ChatGPT) as a learning aid, examining its effects on engagement, usability, and educational achievement. Using the technology acceptance model (TAM) as the conceptual framework, a 24-item survey was designed and administered through simple random sampling. Results reveal that students largely view ChatGPT positively, describing it as an accessible and beneficial tool that promotes flexible and independent learning. The findings confirm TAM predictions, showing that perceived ease of use and favorable attitudes significantly influence adoption. By combining insights from both distance and on-campus learners, the study enriches understanding of how generative artificial intelligence enhances cognitive development and learning adaptability. Overall, the research reflects cautious optimism about ChatGPT’s pedagogical value, acknowledging both its educational benefits and challenges while underscoring the essential role of usability and learner attitudes in fostering sustained artificial intelligence utilization in higher education.
Volume: 15
Issue: 3
Page: 921-931
Publish at: 2026-09-01

Effect of post-weld heat treatment and cooling techniques on mechanical performance of AH36 steel for marine structures

10.11591/ijaas.v15.i3.pp944-953
Ahmad Firdhaus , Blinka Hernawan Prasetya , Muhammad Adnan Rivaldi , Parlindungan Manik , Tuswan Tuswan
The mechanical properties of structural steels used in marine environments are significantly affected by welding procedures. This research examines the impact of various cooling techniques, coupled with post-weld heat treatment (PWHT), on the strength and ductility of AH36 steel welded via shielded metal arc welding (SMAW). Heat treatments were conducted at temperatures of 450 °C, 600 °C, and 750 °C. Oil quenching and normalizing were employed as the two cooling processes. Tensile strength, elastic modulus, elongation, and impact toughness were measured through mechanical testing. The findings demonstrate that oil quenching after treatment at 450 °C yields the most favorable tensile strength and elongation properties, surpassing established safety standards. Conversely, increasing the treatment temperature to 600 °C and 750 °C reduces strength and ductility, presumably due to grain coarsening and microstructural weakening. Impact toughness is substantially improved by normalizing, especially at 600 °C, which offers the best balance of strength and toughness. These findings show that PWHT at 600 °C followed by normalizing markedly improves the mechanical properties of AH36 steel, ensuring dependable performance in harsh marine environments. This research helps improve weld quality and the safety of marine structures by carefully selecting heat treatment and cooling procedures.
Volume: 15
Issue: 3
Page: 944-953
Publish at: 2026-09-01

Rhino optimizer for hyperparameter tuning in depression detection

10.11591/ijaas.v15.i3.pp1239-1251
Simisani Ndaba , Rajalakshmi Selvaraj , Hlomani Hlomani
Depression detection using machine learning (ML) has gained increasing attention; however, model performance remains highly sensitive to hyperparameter configuration. This study introduces the Rhino optimizer, a novel bio-inspired metaheuristic algorithm for hyperparameter optimization (HPO) inspired by behavioral mechanisms observed in Rhinoceros movement and survival strategies. The optimizer integrates Lévy-flight exploration, elite-guided exploitation, wallowing-based escape behavior, and migration-based repositioning to balance exploration and exploitation while reducing premature convergence. The proposed approach was evaluated on a transformer-based depression-detection model using F1-score as the primary optimization objective due to the class imbalance commonly present in mental-health datasets. Performance was compared against traditional search methods, classical metaheuristics, and modern bio-inspired optimization algorithms under an identical evaluation budget. Rhino optimizer achieved a best observed F1-score of 0.8005, equivalent to the sine-cosine algorithm (SCA) and grey wolf optimization (GWO) with the same evaluation budget. Consistent convergence behavior and statistical analyses suggested that the proposed Rhino optimizer may serve as a competitive approach for HPO in depression detection and that behavior-inspired optimization strategies might be useful in machine-learning applications with limited computational resources.
Volume: 15
Issue: 3
Page: 1239-1251
Publish at: 2026-09-01

Mapping significant clinical factors to relapse-free survival prognosis in breast cancer patients

10.11591/ijaas.v15.i3.pp1193-1202
Boby Al Qurthuby , I Made Murwantara
This study explores the role of clinical factors in influencing relapse-free survival (RFS) in breast cancer patients through comprehensive survival analysis (SA). The data were derived from a large cohort of breast cancer patients, encompassing clinical variables such as age at diagnosis, tumor size, number of positive lymph nodes, tumor grade, histological subtype, and receipt of chemotherapy and radiotherapy. Analytical methods included Kaplan-Meier plots, log-rank tests, Cox proportional hazards modeling, and feature ranking based on univariate and multivariate log-rank and Cox tests. The results show that patients who received chemotherapy demonstrated a shorter median RFS compared to those who did not (102.7 vs. 145.0 months, p = 0.03), likely due to high-risk patients being selected for chemotherapy. Meanwhile, radiotherapy was associated with improved prognosis (hazard ratio (HR) < 1 in Cox model, p < 10−5), in line with meta-analyses showing reduced breast cancer mortality post-radiotherapy. The contribution maps the key clinical determinants of RFS in breast cancer and highlights the practicality of performing SA.
Volume: 15
Issue: 3
Page: 1193-1202
Publish at: 2026-09-01
1 2 3
...
2091 2092
Show 25 of 2092

Discover Our Library

Embark on a journey through our expansive collection of articles and let curiosity lead your path to innovation.

Explore Now
Library 3D Ilustration