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

29,196 Article Results

Design and construction of an Arduino-based baby incubator simulator using IoT

10.11591/ijeecs.v41.i1.pp99-108
Liza Rusdiyana , Joel Juanda Jamot Damanik , Bambang Sampurno , Suhariyanto Suhariyanto , Mahirul Mursid , Ika Silviana Widianti
This study aims to create a baby incubator simulator equipped with an internet of things (IoT)-based temperature control system using Arduino UNO. We use a DHT22 sensor to measure temperature and humidity, as well as fuzzy logic to ensure more accurate and responsive temperature control. The Thinger.io platform enables real-time monitoring and control of the incubator, providing flexibility and ease of supervision. With fuzzy logic, the temperature control system can handle changes and uncertainties in the incubator environment, providing a smoother response compared to traditional on-off methods. Testing shows that this system has a very low error rate, with an error value of only 0.97%, meaning that the measured temperature is almost identical to the actual conditions inside the incubator. Additionally, the authors used mice as a model for premature infants in the testing. The results showed that the mice's body temperature increased gradually and stably in line with the incubator conditions, reaching the desired temperature within 90 minutes. This demonstrates that our temperature control system is capable of maintaining optimal environmental conditions for premature infants.
Volume: 41
Issue: 1
Page: 99-108
Publish at: 2026-01-01

Weighted fine-tuned BERT-based sparse RNN for fake news detection

10.11591/ijeecs.v41.i1.pp331-343
Asha Kathigi , Gautam Vinod Nair , Kruthika Kadurahalli Raghu , Kavya Pujar Prakash , Meghana Deepak Duttargi
Fake news refers to misinformation or false reports shared in the form of images, articles, or videos that are disguised as real news to try to manipulate people’s opinions. However, detection systems fail to capture diverse features of fake news due to variability in linguistic styles, contexts, and sources, which lead to inaccurate identification. For this purpose, a weighted fine-tuned-bidirectional encoder representation for transformer based sparse recurrent neural network (WFT-BERT-SRNN) is proposed for fake news detection using deep learning (DL). Initially, data is acquired from Buzzfeed PolitiFact, Fakeddit, and Weibo datasets to evaluate WFT BERT-SRNN. Pre-processing is established using stopword removal, tokenization, and stemming to eliminate unwanted phrases or words. Then, WFT-BERT is employed to extract features. Finally, SRNN is employed to detect and classify fake news as real or fake. Existing techniques like deep neural networks for Fake news detection (DeepFake), BERT with joint learning, and multi-EDU structure for Fake news detection (EDU4FD), Image caption-based technique, and fine-grained multimodal fusion network (FMFN) are compared with WFT-BERT-SRNN. The WFT-BERT-SRNN achieves a better accuracy of 0.9847, 0.9724, 0.9624, and 0.9725 for Buzzfeed, Politifact, Fakeddit, and Weibo datasets compared to existing techniques like DeepFake, BERT-joint framework, EDU4FD, Image caption-based technique, and FMFN.
Volume: 41
Issue: 1
Page: 331-343
Publish at: 2026-01-01

A hybrid divisive K-means framework for big data–driven poverty analysis in Central Java Province

10.11591/ijeecs.v41.i1.pp258-269
Bowo Winarno , Budi Warsito , Bayu Surarso
Clustering is essential in big data analytics, especially for partitioning high dimensional socioeconomic datasets to support interpretation and policy decisions. While K-Means is widely used for its simplicity and scalability, its strong sensitivity to initial centroid selection often leads to unstable results and slower convergence. Previous hybrid approaches, such as Agglomerative–K-Means, attempted to address this issue by using hierarchical clustering for centroid initialization; however, these methods rely on bottom-up merging, which can produce suboptimal initial partitions and increase computational overhead for larger datasets. To overcome these limitations, this study proposes a hybrid divisive–K-Means (DHC) model that employs top-down hierarchical splitting to generate more coherent initial centroids before refinement with K-Means. Using a multidimensional poverty dataset from Central Java Province provided by the Indonesian Central Bureau of Statistics (BPS), the performance of DHC was evaluated against standard K-Means and Agglomerative–K-Means. The assessment included execution time, convergence iterations, and cluster validity indices (Silhouette, Davies–Bouldin, and Calinski–Harabasz). Experimental results demonstrate that DHC reduces execution time by up to 97% and requires 40% fewer iterations than standard K-Means, while achieving comparable or improved cluster quality (e.g., CH Index increasing from 14.3 to 15.8). These findings indicate that the DHC model offers a more efficient and stable clustering solution, addressing the shortcomings of previous standard K-Means methods and improving performance for large-scale socioeconomic data analysis.
Volume: 41
Issue: 1
Page: 258-269
Publish at: 2026-01-01

Evaluation of the performance of mobile telephone networks: literature review

10.11591/ijeecs.v41.i1.pp128-139
Pascal Valandi , Djorwe Temoa , Nsouandele Jean Luc , Tsama Eloundou Pascal , Dokrom Froumsia
Improving the quality of service (QoS) of telephone networks inevitably involves studying previous work on the evaluation of its performance indicators. Several researchers have addressed the subject of evaluating the performance of service of mobile telephone networks. Some proceeded through user surveys and others opted for more objective methods using either professional scanners or developed: hyper text markup language (HTML) or Android applications. The results show that whether by subjective or objective methods, this work has made it possible to advance research and allow other researchers to progress further in the process of evaluating mobile networks. In this study which constitutes a review of the literature, we investigated the different approaches, methods, and most recent results mentioned by researchers to evaluate the QoS by relying much more on objective evaluation. Despite the advances and their limits, in our proposal we intend to rely on data sciences through their tools to evaluate the QoS with more precision.
Volume: 41
Issue: 1
Page: 128-139
Publish at: 2026-01-01

Cyber physical systems maintenance with explainable unsupervised machine learning

10.11591/ijeecs.v41.i1.pp300-308
V. Durga Prasad Jasti , Koudegai Ashok , Ramarao Gude , Prabhakar Kandukuri , Surendra Nadh Benarji Bejjam , Anusha B.
As cyber-physical systems (CPS) continue to play a pivotal role in modern technological landscapes, the need for robust and transparent machine learning (ML) models becomes imperative. This research paper explores the integration of explainable artificial intelligence (XAI) principles into unsupervised machine learning (UML) techniques for enhancing the interpretability and understanding of complex relationships within CPS. The key focus areas include the application of self-organizing maps (SOMs) as a representative unsupervised learning algorithm and the incorporation of interpretable ML methodologies. The study delves into the challenges posed by the inherently intricate nature of CPS data, characterized by the fusion of physical processes and digital components. Traditional black-box approaches in unsupervised learning often hinder the comprehension of model-generated insights, making them less suitable for critical CPS applications. In response, this research introduces a novel framework that leverages SOMs, a powerful unsupervised technique, while concurrently ensuring interpretability through XAI techniques. The paper provides a comprehensive overview of existing XAI methods and their adaptation to unsupervised learning paradigms. Special emphasis is placed on developing transparent representations of learned patterns within the CPS domain. The proposed approach aims to enhance model interpretability through the generation of human-understandable visualizations and explanations, bridging the gap between advanced ML models and domain experts.
Volume: 41
Issue: 1
Page: 300-308
Publish at: 2026-01-01

Convolutional neural network DenseNet in classifying dyslexic handwriting images

10.11591/ijeecs.v41.i1.pp220-232
Chelsea Zaomi Pondayu , Widodo Widodo , Murien Nugraheni
Dyslexia is a specific learning disability (SLD) associated with word-level reading difficulties and often manifests in childhood handwriting through irregular spacing and inconsistent letter sizing, due to shared phonological and orthographic processing. Early identification is critical; however, traditional diagnostic procedures are time-consuming and unsuitable for large-scale screening. This study aimed to develop a handwriting analysis at the paragraph-level using a DenseNet121 convolutional neural network (CNN) model as a low-cost dyslexia screening tool for resource-constrained educational settings. One hundred English handwriting images were preprocessed and standardized into two hundred samples, with 70% of the dataset evaluated using 4-fold cross-validation and the remaining 30% used for testing. The model achieved 90% test accuracy and 92.86% training accuracy, significantly outperforming a random forest baseline that reached 83.57% train accuracy and 63.33% test accuracy, with statistical significance confirmed by McNemar’s test. The main contribution of this study is the demonstration that a lightweight, single-architecture DenseNet121 using paragraph-level analysis can achieve competitive performance compared to prior studies that relied on more complex hybrid models and character-level analysis, while requiring substantially lower computational resources and simplified pipeline. These findings indicate that DenseNet121 provides a robust and low-cost solution for preliminary dyslexia screening in resource-limited educational environments.
Volume: 41
Issue: 1
Page: 220-232
Publish at: 2026-01-01

Lossy ECG signal compression based on RR intervals detection with wavelet transform and optimized run-length encoding

10.11591/ijeecs.v41.i1.pp109-118
Nabil Boukhennoufa , Messaoud Garah
It is expensive to transmit or store significant amounts of electrocardiogram (ECG) records, particularly when using telecommunications channels that charge according to the volume of transferred data. The advancement of telemedicine renders compressing ECG signals even more necessary. Compression aims to reduce the size of data while maintaining the features of ECG signals. This paper presents a novel strategy for compressing ECG signals based on 3D format conversion. After identifying the RR intervals, we divide the signal into cardiac cycles and proceed with the cut and align process. A 3D discrete wavelet transform (DWT) is employed to minimize the correlation existing between two adjacent voxels. Moreover, an optimized run-length encoding (RLE), a novel lossless compression technique, has been proposed to increase the compression ratio (CR). The proposed strategy is applied to different types of ECG records of the Arryyhmia database. This algorithm demonstrates improved performance in terms of CR and percentage root-mean-square difference (PRD) compared to several recently published works.
Volume: 41
Issue: 1
Page: 109-118
Publish at: 2026-01-01

Ultra-high isolation dual-port circular patch antenna at 2.4 GHz

10.11591/ijeecs.v41.i1.pp140-152
Meriem Boucif , Fayza Bousalah , Hayat Benosman
Reliable wireless communication in the 2.4 GHz industrial, scientific, and medical band increasingly relies on antenna systems that can provide high inter-port isolation in multiple-input multiple-output (MIMO) configurations. This paper presents a circular microstrip patch antenna and its extension to a dual-port MIMO configuration designed for 2.4 GHz operation. The antenna is implemented on a low-loss substrate and evaluated using full-wave electromagnetic simulations to assess impedance matching, radiation performance, and MIMO diversity metrics. To enhance inter-port isolation in the array, an inverted U-shaped defected ground structure (DGS) is introduced between the two radiating elements. The optimized design achieves excellent matching around 2.4 GHz and ultra-high isolation of approximately -78.7 dB, while maintaining stable gain and radiation patterns across the operating band. These results indicate that the proposed antenna offers a simple and effective solution for compact, energy-efficient, and robust 2.4 GHz MIMO front ends in internet of things (IoT) and other shortrange wireless communication systems.
Volume: 41
Issue: 1
Page: 140-152
Publish at: 2026-01-01

Comparative analysis of linear regression, random forest, and LightGBM for hepatitis disease prediction

10.11591/ijeecs.v41.i1.pp430-438
Hennie Tuhuteru , Goldy Valendria Nivaan , Marvelous Marvel Rijoly , Joselina Tuhuteru
In bioinformatics research, computational pattern-analysis techniques are frequently employed to assist in disease prediction and diagnostic modeling, including applications for hepatitis prognosis. Hepatitis is a type of serious disease with various types that have the potential to threaten the life of the sufferer without showing significant symptoms and signs, so many sufferers do not realize that they are affected by the disease. Various methods are used to predict diseases in the hope of providing the best results from the learning model used. The objective of this study is to implement linear regression, random forest, and light gradient boosting machine (LightGBM) to estimate mortality risk among hepatitis patients. In addition, a performance comparison of the results of hepatitis disease prediction using the three algorithms was also carried out to find out which model gave the most accurate and optimal results. The results of this study show that the application of learning models from the linear regression, random forest and Light-GBM algorithms has been successfully carried out to predict the survival status of patients with hepatitis. The findings reveal that random forest achieved the highest predictive performance with an accuracy of 84%, followed by LightGBM at 77% and linear regression at 32%.
Volume: 41
Issue: 1
Page: 430-438
Publish at: 2026-01-01

YOLOv8m enhancement using α-scaled gradient-normalized sigmoid activation for intelligent vehicle classification

10.11591/ijeecs.v41.i1.pp153-167
Renz Raniel V. Serrano , Jen Aldwayne B. Delmo , Cristina Amor M. Rosales
Vehicle classification plays a vital part in the development of intelligent transportation systems (ITS) and modern traffic management, where the ability to detect and identify vehicles accurately in real time is essential for maintaining road efficiency and safety. This paper presents an enhancement to the YOLOv8m model by refining its activation function to achieve higher accuracy and faster response in diverse traffic and environmental situations. In this study, two alternative activation functions—Mish and Swish—were integrated into the YOLOv8m structure and tested against the model’s default sigmoid linear unit (SiLU). Training and evaluation were carried out using a comprehensive dataset of vehicles captured under different lighting and weather conditions. The experimental findings show that the modified activation design leads to better model convergence, improved generalization, and a noticeable boost in detection performance, recording up to 5.4% higher accuracy and 6.6% better mAP scores than the standard YOLOv8m. Overall, the results confirm that fine-tuning activation behavior can make deep learning models more adaptive and reliable for vehicle classification tasks in real-world intelligent transportation environments.
Volume: 41
Issue: 1
Page: 153-167
Publish at: 2026-01-01

Cryptojacking detection using model-agnostic explainability

10.11591/ijeecs.v41.i1.pp394-408
Elodie Ngoie Mutombo , Mike Wa Nkongolo , Mahmut Tokmak
Cryptojacking is the illicit use of computing resources for cryptocurrency mining. It has emerged as a serious cybersecurity threat that degrades critical system performance and increases operational costs. This paper proposes an advanced machine learning (ML) framework that integrates transformer based language models with post hoc explainable artificial intelligence (XAI) to detect cryptojacking using complementary network traffic and process memory data. Numerical and categorical features are discretized and tokenized to enable semantic modelling and contextual learning. Experimental results show that transformer models effectively capture cryptojacking-related behavioral patterns, with decoding-enhanced BERT with disentangled attention (DeBERTa) achieving high detection performance and recall exceeding 80%. bidirectional encoder representations from transformers (BERT) attains comparable recall with lower computational overhead, making it well suited for real-time environments, while robustly optimized BERT approach (RoBERTa) and DeBERTa are more appropriate for offline or batch-based analysis. Model performance is evaluated using standard classification metrics, and XAI techniques provide interpretable insights into feature relevance, supporting transparent and reliable detection. In general, the proposed framework delivers an effective and deployment-ready solution for cryptojacking detection.
Volume: 41
Issue: 1
Page: 394-408
Publish at: 2026-01-01

Relationship between voltage and resistance in hybrid nanoconductive ink on different substrates in wet and dry conditions

10.11591/ijeecs.v41.i1.pp18-32
Norashikin Shari , Nurfaizey Abd Hamid , Chonlatee Photong , Alan J. Watson , Mohd Azli Salim
Hybrid graphene nanoplatelet/silver (GNP/Ag/SA) conductive inks are increasingly used in flexible electronics, yet there is limited understanding of how substrate type, solvent composition, and moisture exposure jointly control the electrical performance on metal and polymer substrates. This work aims to clarify how terpinol content (5T, 10T, 15T) and substrate properties of copper (Cu), polyethylene terephthalate (PET), and thermoplastic polyurethane (TPU) influence voltage, resistance, and resistivity of screen-printed GNP/Ag/SA tracks under dry and postimmersion wet conditions. GNP/Ag/SA inks were formulated with fixed butanol and varied terpinol contents, printed on Cu, PET, and TPU, and characterized using electrical measurements, adhesion evaluation, and microstructural observations to relate resistivity trends to morphology, surface energy, and hygroscopic behavior. The Cu substrate showed the best performance, with Cu 10T achieving the lowest dry resistivity of approximately 1.2×10-5 Ω.m and Cu 15T the lowest wet resistivity of approximately 2.0×10-5 Ω.m, supported by dense, well-adhered microstructures. The PET exhibited higher resistivity values up to about 10-3 Ω.m and clear degradation after water immersion, while TPU showed very high or unmeasurable resistivity in wet conditions caused by severe ink loss and hygroscopic swelling, highlighting the important role of substrate surface energy and moisture response in determining the reliability of GNP/Ag/SA inks for applications in humid or wet environments.
Volume: 41
Issue: 1
Page: 18-32
Publish at: 2026-01-01

Development of unified college admission system for Philippine state universities and colleges: a data-driven approach to equity and access

10.11591/ijeecs.v41.i1.pp61-72
Abegail G. Bordios , Sherrie Ann Cananua-Labid , Ariel B. Mabansag , Mae V. Cañal , Jake Boy D. Carboquillo , Ma. Andrea C. Del Rosario
This paper presents the development and pilot evaluation of the unified college admission system (UCAS), a centralized and equity-oriented digital platform designed to streamline admissions across Philippine state universities and colleges (SUCs). Anchored on Republic Act No. 10931, UCAS functions as a unified application repository that standardizes admissions data, consolidates applicant records, and enables real-time monitoring of equity target students (ETS) to support fair and transparent access to higher education. The system integrates student-facing and administrative portals that facilitate application submission, institutional coordination, and equity-focused analytics. A pilot evaluation involving student applicants and administrators assessed usability, efficiency, and reliability, yielding consistently positive results across user groups. Findings indicate that UCAS is technically robust, user-centered, and suitable for multi-level admissions governance. Overall, the study demonstrates the potential of a centralized, data-driven admissions platform to complement tuition-free education policies by addressing inequities at the admissions stage.
Volume: 41
Issue: 1
Page: 61-72
Publish at: 2026-01-01

Lung cancer segmentation and classification using hybrid CNN-LSTM model

10.11591/ijeecs.v41.i1.pp309-319
Manaswini Pradhan , Ahmed Alkhayyat
A collection of genetic disorders and various types of abnormalities in the metabolism lead to cancer, a fatal disease. Lung and colon cancer are found to be main causes of death and infirmity in people. When choosing the best course of treatment, the diagnosis of these tumors is usually the most important consideration. This study's main objectives are to classify lung cancer and its severity, as well as to recognize malignant lung nodules. The suggested approach additionally classifies the stages of lung cancer in order to recognize lung nodules. The convolutional neural network (CNN) is used to detect lung nodules, identifying a nodule which is accurately segmented and classified. The suggested method is separated into dual parts: primarily, it classifies normal and abnormal behavior, and the subsequent one classifies the different stages of lung cancer. Texture and intensity-based features are extracted during the classification stage. When compared to other methods such as nested long short-term memory (LSTM)+ CNN, the hybrid CNN LSTM obtains superior outcomes in terms of accuracy (99.35%), specificity (99.30%), sensitivity (99.32%), and F1-score (99.29%).
Volume: 41
Issue: 1
Page: 309-319
Publish at: 2026-01-01

A novel approach for detecting diabetic retinopathy using two-stream CNNs model

10.11591/ijeecs.v41.i1.pp200-209
Pham Thi Viet Huong , Le Duc Thinh , Tran Thi Oanh , Tran Xuan Bach , Hoang Quang Huy , Tran Anh Vu
Major causes of visual impairment, particularly diabetic retinopathy (DR) and aged-related macular degeneration (AMD), has posed significant challenges for clinical diagnosis and treatment. Early detection and prompt intervention can help prevent severe consequences for patients. The study presents a novel approach for detecting eye diseases using a two-stream convolutional neural network (CNN) model. The first stream processes preprocessed fundus images, while the second stream analyzes high-pass filtered fundus images in the spatial frequency domain. To assess the model’s performance, we use the APTOS 2019 dataset, which was originally compiled for the Asia Pacific Tele-Ophthalmology Society 2019 Blindness Detection competition and is publicly available on Kaggle. Our method shows promise as an early screening tool for DR detection with an accuracy of 0.986.
Volume: 41
Issue: 1
Page: 200-209
Publish at: 2026-01-01
1 2 3
...
1946 1947
Show 9 of 1947

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