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

29,758 Article Results

Analysis of railway accidents in Nigeria: a decade of insights

10.11591/ijaas.v15.i1.pp19-28
Aliyu Mani Umar , Mohd Khairul Afzan Mohd Lazi , Sitti Asmah Hassan , Hanini Ilyana Che Hashim , Yinggui Zhang , Nura Shehu Aliyu Yaro , Adam Ado Sabari , Surajo Abubakar Wada
This study provides insights into the patterns and dynamics of railway accidents in Nigeria over the past decade. Findings indicate that Nigeria's rail network experiences fewer but more severe accidents than the United States of America (USA) and United Kingdom (UK), with significantly higher fatalities and injuries per million train kilometers 92% and up to 95% more, respectively, in 2023. A top-down approach was employed to establish a risk tree, revealing six railway accident categories recorded over the last decade. The established risk tree could provide a framework for conducting the rail network's comprehensive safety risk assessment. Finally, a root cause analysis of railway intrusion accidents, the most occurring railway accident category in the Nigerian rail network, was conducted. Six immediate and eleven underlying causes (UC) of railway intrusion accidents were identified. About 62% of all intrusion accidents were caused by negligence of road users. Several actionable preventive measures (PM) have been proposed for each identified UC based on best practices in developed rail networks. Infrastructure upgrades and safety awareness campaigns have been identified as the potentially most effective PM for railway intrusion accidents in Nigeria.
Volume: 15
Issue: 1
Page: 19-28
Publish at: 2026-03-01

SAIDI and SAIFI indicators for the control of feeder A4502 of the Concepción transformer electrical substation

10.11591/ijaas.v15.i1.pp396-404
Margarita F. Murillo Manrique , Jorge Augusto Sánchez Ayte , William Joel Baygorrea Vega , Richard Flores-Caceres
This study evaluated the reliability of feeder A4502 of the Concepción substation (Huancayo, Peru) through the analysis of system average interruption duration index (SAIDI) and system average interruption frequency index (SAIFI) indicators. The 46-year-old infrastructure presented 805 structural deficiencies (59%), with a predominance of corrosion in iron poles. Automatic recloser devices were implemented at strategic points, based on the fact that 67% of the 73 interruptions in 2021 were transient faults. Post-intervention results (2024) showed significant improvements: SAIDI was reduced from 9.87 to 7.39 hours (25%), nearing the regulatory limit of 7 hours; SAIFI decreased from 4.29 to 2.71 events (37%), falling within the limit of 4. Pearson correlation analysis confirmed a statistically significant relationship between structural deficiencies and the indicators (r =0.62 SAIDI, r =0.58 SAIFI, p <0.05). The integrated approach—diagnosis of deficiencies + automation with reclosers—proved to be technically viable and economically justifiable, also allowing for the meeting of new energy demands (240 kVA available). The results constitute a replicable model for other aging Latin American networks, validating the viability of regulatory compliance without prohibitive investments.
Volume: 15
Issue: 1
Page: 396-404
Publish at: 2026-03-01

Hybrid deep learning and ensemble learning approach for high accuracy thyroid disease classification

10.11591/ijaas.v15.i1.pp303-312
Shuriya Balusamy , Balajishanmugam Vivekanadhan , Prathima Mabel John , Sushma Sunil Bhosle
Thyroid disease is a common endocrine disorder affecting the thyroid gland, a small butterfly-shaped organ at the base of the neck. According to the World Health Organization (WHO), nearly one billion people worldwide are affected by thyroid-related conditions. Conventional diagnostic methods, such as thyroid scans and function tests, are often costly, time-consuming, and complex for clinicians to interpret. To overcome these limitations, this study introduces a novel temporal conditional-Markov random field (TC MRF) framework for early detection and classification of thyroid disease. The multi-modality images computed tomography (CT), magnetic resonance imaging (MRI), and ultrasound (US) are collected from the ImageNet database and preprocessed using contrast stretching adaptive Gaussian star (CSAGS) filter to improve image clarity. The enhanced images are then processed over a convolutional neural network (CNN) for feature extraction. These features are classified using a random forest (RF) model to determine whether the thyroid condition is normal or abnormal. The proposed TC MRF achieves a classification accuracy of 98.27% and F1-score of 96.05%. The TC-MRF enhances the total accuracy range of 6.30%, 4.11%, and 5.36% better than naive Bayes, multilayer perceptron (MLP), and decision tree, respectively.
Volume: 15
Issue: 1
Page: 303-312
Publish at: 2026-03-01

Robust multi-faces recognition and tracking via fuzzy genetic algorithms and deep coupled features

10.11591/ijaas.v15.i1.pp209-218
Adil Abdulhur Abushana , Yousif Samer Mudhafar
In real-world surveillance environments, face recognition and tracking remain challenging due to partial occlusion, pose variation, illumination changes, and background clutter. This paper presents a robust hybrid framework that integrates fuzzy genetic algorithms (FGA) with deep coupled feature learning for multi-face recognition and tracking. The proposed system comprises three main modules: i) face detection and pre processing using the multi-task cascaded convolutional network (MTCNN), ii) deep coupled ResNet embeddings that jointly learn identity and appearance-invariant representations, and iii) a fuzzy rule-based genetic optimizer that adaptively refines tracking decisions based on uncertainty in motion, appearance similarity, and occlusion levels. The novelty of this work lies in the fusion of fuzzy inference with evolutionary search to guide the genetic optimization process—allowing dynamic adaptation to noisy and uncertain visual conditions. Moreover, probabilistic data association filters (PDAF) and conditional joint likelihood filters (CJLF) are employed to further enhance temporal consistency under occlusion and appearance variation. The results confirm that fuzzy evolutionary optimization, when coupled with deep feature learning, significantly improves robustness and stability for real-time face tracking in complex, dynamic scenes.
Volume: 15
Issue: 1
Page: 209-218
Publish at: 2026-03-01

Experimental study on annealing S45C steel: effect of temperature and time on hardness, impact strength

10.11591/ijaas.v15.i1.pp343-354
Mahadir Sirman , Syahrisal Syahrisal , Henny Pasandang , Rusdi Nur , Muhira Dzar Faraby , Mukhlisin Mukhlisin
Steel generally exhibits poor wear and friction resistance, making it necessary to improve its surface mechanical properties, particularly hardness and microstructure, to enhance performance. Heat treatment is one of the most effective methods for achieving these improvements. This study aimed to optimize the heat treatment parameters of S45C medium-carbon steel to improve hardness and impact strength using response surface methodology (RSM). Experimental trials were conducted at annealing temperatures of 800 °C, 850 °C, and 900 °C with holding times of 30, 60, and 90 minutes, followed by cooling in water, oil, or air. Hardness (HRC) and impact strength (Nm/mm²) were measured, and the data were analyzed using RSM with a central composite design (CCD). Quadratic models were found to be statistically significant for both hardness (Prob > F = 0.0222) and impact strength (Prob > F = 0.0338), confirming their validity. The optimization results indicated that a holding time of 60 minutes within the 850-900 °C range provides the best balance between high hardness (>55 HRC) and adequate impact strength (>0.68 Nm/mm²). These findings not only validate the predictive capability of RSM in heat treatment optimization but also provide practical guidelines for industrial applications of S45C steel in automotive, tooling, and structural components.
Volume: 15
Issue: 1
Page: 343-354
Publish at: 2026-03-01

Corporate social responsibility by listed commercial banks in Vietnam: practice and financial performance

10.11591/ijaas.v15.i1.pp261-271
Viet Ha Nguyen , Thi Minh Nguyet Dang
This study examines the impacts of financial performance (FP) on corporate social responsibility (CSR). The article investigates whether FP, as measured by return on assets (ROA) and net interest margin (NIM), influences the likelihood of CSR disclosure, drawing on stakeholder theory and legitimacy theory. The analysis employs binary logistic regression models and an unbalanced panel dataset comprising 26 listed banks between 2014 and 2024. If the bank discloses its CSR practices in its annual or sustainability report, the code for CSR disclosure is 1; otherwise, it is 0. The results show that, while NIM shows a negative relationship, ROA significantly improves CSR. Furthermore, there is a positive correlation between bank size (TA), equity to asset (EA), and CSR; a negative relationship of loan to deposit ratio (LDR) with CSR, and no significant statistical correlation was observed between debt to equity (DTE) and CSR. The study adds to the body of knowledge on CSR in developing nations and offers recommendations for sustainability and bank governance.
Volume: 15
Issue: 1
Page: 261-271
Publish at: 2026-03-01

An ensemble-based approach for breast cancer identification using mammography

10.11591/ijaas.v15.i1.pp133-141
Naveen Ananda Kumar Joseph Annaiah , Nakka Thirupathi Rao , Balakesava Reddy Parvathala , Banana Omkar Lakshmi Jagan , Bodapati Venkata Rajanna
Breast cancer is among the most common cancers in women worldwide; timely detection is vitally important for improving chances of survival. The present study examines an innovative machine learning technique for the diagnosis of breast cancer using the breast cancer Wisconsin (diagnostic) dataset from Kaggle. The dataset includes 569 instances, and each instance has 30 attributes derived from digitized fine needle aspiration (FNA) images of masses found in the breast. We will present an ensemble deep learning (DL) model fusing a convolutional neural network (CNN) and LRAlexNet architectures to increase the accuracy and robustness of this type of cancer diagnosis. CNN models are well-known for their power to capture spatial hierarchies in image data, and LRAlexNet is a specialized deep CNN that excels at image classification due to its depth and parameter optimization. In this work, we use the ability to extract features of CNNs along with the superior classification performance of LRAlexNet to distinguish between benign and malignant cancers. The model will be trained and validated on the curated breast imaging subset of the digital database for screening mammography (CBIS-DDSM) dataset, and performance will be evaluated using sensitivity, accuracy, specificity, and the area under the curve (AUC) for the receiver operating characteristic. The results show that the ensemble CNN-LRAlexNet model achieved superior accuracy for breast cancer prediction when compared to traditional machine learning methods.
Volume: 15
Issue: 1
Page: 133-141
Publish at: 2026-03-01

Effectiveness of iBreast examination for screening breast lesions among women in India

10.11591/ijaas.v15.i1.pp178-186
Samuel Ani Grace Kalaimathi , Venkatesan Hemavathy , Sambavadas Kanchana , Radhakrishnan Sudha , Perumal Tamilarasi
The breast has long been a representation of women's identity and an essential component of fertility. The breast lesions refer to an area of abnormal breast tissue. One frequent medical ailment that might worry women is breast lesions. It is estimated that at least 20% of females may develop breast lesions. It may vary in size, shape, and texture can be either benign or malignant. Mammography, clinical breast examination (CBE), and self-breast inspection are the accepted early breast cancer detection techniques. Mammography application in low and middle-income countries is limited because most of the women in these countries cannot afford it. Hence, iBreastExam was identified and validated as an alternative source for screening at the village level to identify breast lesions at an early stage. For the study, a cross-sectional research design using a quantitative research methodology was used. Adopted areas of the selected colleges were the setting for the study: MA Chidambaram College of Nursing, Adyar, Chennai; Sri Balaji College of Nursing, Chrompet, Chennai; Madha College of Nursing, Kundrathur, Chennai; Omayal Achi College of Nursing, Puzhal, Chennai. The sample size consisted of 14,000 women across all the 4 settings. A convenient sampling technique was used to select the samples for the study. A total of 13,988 women were screened, 55 women had positive breast lesions, and out of this 5 were confirmed to have breast cancer through mammogram diagnosis.
Volume: 15
Issue: 1
Page: 178-186
Publish at: 2026-03-01

Technical proposal for the design of a helical conveyor for solid waste handling

10.11591/ijaas.v15.i1.pp333-342
Javier Sinche Ccahuana , Jorge Augusto Sánchez Ayte , Margarita F. Murillo Manrique , Richard Flores-Cáceres
The novelty of this work lies in the design of a helical conveyor for solid waste from the chocolate industry, materials that can be cohesive, with variable density, and potentially corrosive. The objective is to present a validated and replicable technical model that optimizes the transport of 5 metric tons per hour of these wastes at Peru's National Chocolate Company. The goal is to minimize human contact, improve ergonomic safety, and transform waste into exploitable resources under circular economy principles. The methodology employed is an applied type with a quantitative approach, supported by the selection of components through specialized technical catalogs from KWS manufacturing and Martin engineering, which implement ANSI/CEMA 350 standards. Results indicate a total required power of 1.5 HP, with a helicoid diameter of 9", a helical tube of 2", a pitch of 6", and operation at 60 RPM. It is concluded that this design constitutes an efficient and replicable technical solution to improve working conditions in industrial environments, significantly reducing occupational injuries while mitigating environmental impact.
Volume: 15
Issue: 1
Page: 333-342
Publish at: 2026-03-01

A novel circulant matrix-based McEliece framework for secure digital communication

10.11591/ijaas.v15.i1.pp293-302
Ravikumar Inakoti , James Stephen Meka , Padala Venkata Gopala Durga Prasad Reddy
McEliece cryptosystem is old and well-explored post-quantum cryptography system that offers superior security against quantum attacks. Though the system holds great potential and superior security, the challenge associated with large key sizes has made system impractical for most applications. The first challenge against McEliece cryptosystem remains its large key sizes, which make system impractical, especially when implementing internet of things (IoT) and mobile communication applications. Overcoming challenges and retaining superior security still remains an issue to explore. This paper presents investigation into use of circulant matrices for McEliece encryption system to achieve a considerable reduction in key sizes and enhance fast encryption processes. The use of circulant matrices’ inherent properties boosts performance without focusing much on system’s security. In addition, the paper presents security evaluation process for modified communication system to determine and mitigate weaknesses that might arise, considering use of sophisticated encryption systems. Findings and results explore use of circulant matrices, which achieve great reductions in key sizes and improve efficiency of process. Security evaluation reports that proper scrambling techniques are efficient at mending the vulnerabilities associated with circulant matrix structures. A modified McEliece cryptosystem using circulant matrices offers superior data communication, balancing both strong security and efficient computational processes, making system ideal for use in recent communication systems.
Volume: 15
Issue: 1
Page: 293-302
Publish at: 2026-03-01

Improved seizure detection using optimized time sequence based deep learning framework

10.11591/ijaas.v15.i1.pp198-208
Puspanjali Mallik , Ajit Kumar Nayak , Satyaprakash Swain
Epilepsy disease originates due to the presence of disordered neurons, and epilepsy detection stands as a challenging task for neurologists. With recent advances, electroencephalography (EEG)-based analysis is increasingly supported by deep learning and metaheuristic optimization approaches in order to improve the test results. This experiment uses a convolutional neural network (CNN) model hybridized with bidirectional long short-term memory (BiLSTM). CNN leverages the work with improved feature extraction cum classification supports, and BiLSTM keeps the time sequence of data in both the forward and backward direction for improving signal mapping purposes. To reduce the computational overhead and improve execution accuracy, a hybrid optimization algorithm called secretary bird optimization algorithm (SBOA) is used to fine-tune the execution. Key classification parameters such as accuracy, sensitivity, and specificity reflect the model’s strong predictive capability, with accuracy reaching up to 98.49%. The proposed method demonstrates the potential for high-performance EEG-based seizure detection, paving the way for future integration with edge computing devices to support remote clinical diagnostics and continuous monitoring in real-world healthcare applications.
Volume: 15
Issue: 1
Page: 198-208
Publish at: 2026-03-01

Extension of Hermite-Hadamard type inequalities to Katugampola fractional integrals

10.11591/ijaas.v15.i1.pp1-18
Dipak Kr Das , Shashi Kant Mishra , Pankaj Kumar , Abdelouahed Hamdi
In this study, we introduce several new Hermite-Hadamard type general integral inequalities for exponentially (s,m)-convex functions via Katugampola fractional integral. The Katugampola fractional integral is a broader form of the Riemann–Liouville and Hadamard fractional integrals. We utilized the power mean integral inequality, the H¨older inequality and a few additional generalizations to derive these inequalities. Numerous limiting results are derived from the main results presented in the remarks. Furthermore, we provide an example illustrating our theoretical findings, supported by a graphical representation.
Volume: 15
Issue: 1
Page: 1-18
Publish at: 2026-03-01

Performance enhancement of photovoltaic system integrated with a single-phase grid using advanced controllers

10.11591/ijaas.v15.i1.pp77-85
Madhu Babu Thiruveedula , Thiramdasu Chandana , Meghavath Mahesh , Avinash Udala , Yerra Praveen , Mohammed Assaduzzama
This study offers a thorough examination of a photovoltaic (PV) system using a variety of maximum power point tracking (MPPT) methods, including fuzzy logic control (FLC), adaptive neuro-fuzzy inference systems (ANFIS), perturb and observe (P&O), and artificial neural networks (ANN). Optimizing power extraction from PV systems under various environmental circumstances, including temperature variations and irradiance, is the main goal of these MPPT algorithms. Despite its widespread use and affordability, the P&O algorithm may have performance issues in dynamic circumstances. By using fuzzy logic to adjust to non-linear changes in environmental conditions, FLC improves P&O and offers more dependable and seamless operation. Although they demand a large amount of data and processing power, ANN-based MPPT approaches provide sophisticated capabilities by predicting optimal operating points by learning from historical system actions. By fusing fuzzy logic and neural networks, ANFIS offers a reliable solution that can more accurately adjust in real time to changing circumstances. These algorithms' incorporation into a PV system allows for more flexible and effective power management, guaranteeing peak performance in a range of climatic conditions. By combining many MPPT techniques, hybrid approaches can further reduce the drawbacks of individual approaches and improve the overall dependability and efficiency of PV systems.
Volume: 15
Issue: 1
Page: 77-85
Publish at: 2026-03-01

Financial distress prediction for batik small and medium enterprises credit financing based on deep learning algorithm

10.11591/ijaas.v15.i1.pp245-252
Taryadi Taryadi , Bambang Sudiyatno , Robertus Basiya , Era Yunianto
One of the biggest obstacles that any finance provider has when evaluating a borrower's creditworthiness is the prediction of financial trouble. The credit decision-making process is made more difficult for small and medium enterprises (SMEs) due to their inherent ambiguity, which raises financing costs and lowers the chance of approval. In order to estimate a binomial classifier for predicting financial hardship using logistic regression (LR), extreme gradient boosting (XGBoost), and artificial neural network (ANN) techniques, this study examines data from batik SMEs in Pekalongan city. Financial ratios predict the first period and grow in a multi-period model based on temporal factors, credit history, and age. Financial distress is defined as a substantial obstacle to a business's capacity to pay its debts as opposed to the potential for bankruptcy. The long short-term memory (LSTM) algorithm with more variables yields the best prediction accuracy. The study's conclusion indicates that in order to guarantee the accuracy of financial distress prediction, the time at risk must be adjusted.
Volume: 15
Issue: 1
Page: 245-252
Publish at: 2026-03-01

Performance comparison of feature extraction methods for electroencephalogram-based recognition of Balinese script

10.11591/ijaas.v15.i1.pp55-64
I Made Agus Wirawan , Ida Bagus Nyoman Pascima , Gede Surya Mahendra , I Made Candiasa , I Nyoman Sukajaya
Recognizing Balinese script from electroencephalogram (EEG) signals remains a challenging problem due to low signal amplitude, non-stationary dynamics, and significant inter-subject variability. Despite previous attempts, no single feature extraction method has been universally effective in addressing these limitations. To fill this gap, this study systematically evaluates five feature extraction techniques—differential entropy (DE), power spectral density (PSD), discrete wavelet transforms (DWT), Hjorth parameters, and statistical features—on the Balinese imagined spelling using electroencephalography (BISE) dataset, which contains EEG recordings specifically designed for Balinese script recognition. For classification, both artificial neural networks (ANN) and support vector machines (SVM) are applied, and their performance is validated across multiple experimental settings. Results demonstrate that DE consistently provides more stable and discriminative features than the other methods, achieving the highest classification accuracy when combined with ANN. These findings highlight the potential of DE-based approaches to advance electroencephalogram driven Balinese script recognition, offering a culturally significant contribution to brain-computer interface (BCI) research and supporting future applications in inclusive artificial intelligence, digital heritage preservation, and assistive technologies.
Volume: 15
Issue: 1
Page: 55-64
Publish at: 2026-03-01
1 2 3
...
1983 1984
Show 12 of 1984

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