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

29,939 Article Results

Secure financial application using homomorphic encryption

10.11591/ijeecs.v38.i1.pp595-602
Vijaykumar Bidve , Aruna Pavate , Rahul Raut , Shailesh Kediya , Pakiriswamy Sarasu , Koteswara Rao Anne , Aryani Gangadhara , Ashfaq Shaikh
In today’s digital age, the security and privacy of financial transactions are paramount. With the advent of technologies like homomorphic encryption, it is now possible to perform computations on encrypted data without the need to decrypt it first, offering a promising avenue for secure financial applications. This research paper explores the implementation and implications of utilizing homomorphic encryption in financial applications to safeguard sensitive data while maintaining computational integrity. By employing homomorphic encryption techniques, financial institutions can enhance the confidentiality of their clients’ information, protect against data breaches, and enable secure computations on encrypted data. The paper discusses the principles of homomorphic encryption, its applications in financial systems, challenges, and potential solutions. Additionally, it examines real-world examples and case studies where homomorphic encryption has been employed successfully, highlighting its effectiveness in ensuring the privacy and security of financial transactions. Overall, this paper aims to provide insights into the role of homomorphic encryption in creating secure financial applications and its potential to revolutionize the way sensitive financial data is handled and processed.
Volume: 38
Issue: 1
Page: 595-602
Publish at: 2025-04-01

Tree-based models and hyperparameter optimization for assessing employee performance

10.11591/ijeecs.v38.i1.pp569-577
Rendra Gustriansyah , Shinta Puspasari , Ahmad Sanmorino , Nazori Suhandi , Dewi Sartika
The Palembang city fire and rescue service (FRS) is encountering challenges in adhering to national standards for fire response time. Hence, the Palembang city FRS is committed to enhancing employee performance through quarterly performance assessments based on various criteria such as attendance, work targets, behavior, education, and performance reports. This study proposes tree-based models in machine learning (ML) and hyperparameter optimization to assess the performance of Palembang city FRS employees. Tree-based models encompass decision trees (DT), random forests (RF), and extreme gradient boosting (XGB). The predictive performance of each model was evaluated using the confusion matrix (CM), the area under the receiver operating characteristic (AUROC), and the kappa coefficient (KC). The results indicate that RF performs better than DT and XGB in the sensitivity, AUROC, and KC metrics by 1.0000, 0.9874, and 0.8584, respectively.
Volume: 38
Issue: 1
Page: 569-577
Publish at: 2025-04-01

A comprehensive overview of LLM-based approaches for machine translation

10.11591/ijeecs.v38.i1.pp344-356
Bhuvaneswari Kumar , Varalakshmi Murugesan
Statistical machine translation (SMT) used parallel corpora and statistical models, to identify translation patterns and probabilities. Although this method had advantages, it had trouble with idiomatic expressions, context-specific subtleties, and intricate linguistic structures. The subsequent introduction of deep neural networks such as recurrent neural networks (RNNs), long short-term memory (LSTMs), transformers with attention mechanisms, and the emergence of large language model (LLM) frameworks has marked a paradigm shift in machine translation in recent years and has entirely replaced the traditional statistical approaches. The LLMs are able to capture complex language patterns, semantics, and context because they have been trained on enormous volumes of text data. Our study summarizes the most significant contributions in the literature related to LLM prompting, fine-tuning, retrieval augmented generation, improved transformer variants for faster translation, multilingual LLMs, and quality estimation with LLMs. This new research direction guides the development of more efficient and innovative solutions to address the current challenges of LLMs, including hallucinations, translation bias, information leakage, and inaccuracy due to language inconsistencies.
Volume: 38
Issue: 1
Page: 344-356
Publish at: 2025-04-01

TALOS: optimization of the CNN for the detection of the tomato leaf diseases

10.11591/ijeecs.v38.i1.pp292-302
Shruthi Kikkeri Subramanya , Naveen Bettahalli , Naveen Kalenahalli Bhoganna
Early detection of plant diseases using convolutional neural network (CNN)is crucial for maximizing crop yield and minimizing economic losses. Manual inspection, the frequent technique, is inefficient and error prone. While CNN’s offer potential for accurate and quick disease recognition, their performance is highly dependent on effective hyperparameter tuning. This process is time consuming, resource intensive, and needs significant expertise due to the vast hyperparameter space, since it can be hard to figure out which is ideal for optimal performance. An effective optimization tool, tunable automated hyperparameter learning optimization system (TALOS), is proposed, which automates the tuning of hyperparameters by systematically exploring the hyperparameter space and evaluates different combinations of parameters to find the optimal configuration that maximize the model’s performance. The performance of this approach is recognizable through its exploration of five different hyperparameters across a search space of 32 combinations, yielding optimal parameters by the second round. Using 3030 tomato leaf images from a benchmark data set, the model achieves a remarkable 94.7% validation accuracy with 33647 trainable parameters. Thus, automated hyperparameter tuning approach not only optimizes model performance but also reduces manual effort and resource requirements, paving the way for more effective and scalable solutions in agricultural technology.
Volume: 38
Issue: 1
Page: 292-302
Publish at: 2025-04-01

Comparing machine learning models for Indonesia stock market prediction

10.11591/ijeecs.v38.i1.pp508-516
Selly Anastassia Amellia Kharis , Arman Haqqi Anna Zili , Maulana Malik , Wahyu Nuryaningrum , Agustiani Putri
The financial market hold a significant role in the economy and the ability to accurately predict stock prices poses a major challenge, particularly in volatile markets like Indonesia. This study investigates the application of three supervised machine learning algorithms: random forest (RF), support vector regression (SVR), K-nearest neighbor (KNN) to predict the closing prices of stocks. The data used in this research consists of BBCA, PWON, and TOWR stocks. This study adopted daily historical stock prices from March 2017 to February 2020, which were normalized and segmented into training and testing datasets. The models were trained using machine learning techniques, and their predictive accuracy was evaluated using root mean square error (RMSE) and mean absolute error (MAE). The historical stock data includes Open, High, Low, and Close prices. The result indicated that SVR consistently outperforms RF and KNN in terms of RMSE and MAE across different stocks. The SVR method produced RMSE values of 4.79% for BBCA stock, 10.61% for PWON stock, and 15.14% for TOWR stock, and produces MAE values of 3.52% for BBCA stock, 8.49% for PWON stock, and 13.78% for TOWR stock.
Volume: 38
Issue: 1
Page: 508-516
Publish at: 2025-04-01

Enhanced hippopotamus optimization algorithm for power system stabilizers

10.11591/ijeecs.v38.i1.pp22-31
Widi Aribowo , Toufik Mzili , Aliyu Sabo
This article presents techniques for modifying the power system stabilizer's (PSS) parameters. An enhanced version of the hippocampal optimization algorithm (HO) is presented here. HO represents a novel approach in metaheuristic methodology, having been inspired by the observed clinging behavior in hippos. The notion of the HO is defined using a trinary-phase model that includes their position updates in rivers or ponds, defensive techniques against predators, and mathematically described evasive methods. To confirm the efficacy of the recommended approach, this article provides comparison simulations of the PSS objective function and transient response. This study employs validation through a comparison between Original HO and conventional methods. Simulation results demonstrate that, when compared to competing algorithms, the suggested approach yields optimal results and, in some cases, exhibits fast convergence. It is known that, in comparison to the original HO approach, the recommended way can lower the average undershoot of the rotor angel and speed by 12.049% and 26.97%, respectively.
Volume: 38
Issue: 1
Page: 22-31
Publish at: 2025-04-01

Predicting peak demand for electricity consumption using time series data and machine learning model

10.11591/ijeecs.v38.i1.pp668-676
Suriya S. , Agusthiyar R.
Energy consumption is influenced by various factors, including the proliferation of electronic devices, technological advancements, economic growth, agricultural development, and population increase. Each of these factors contributes to the rising demand for energy. This paper addresses the challenge of predicting peak energy demand (ED) by utilizing historical time series data from the past five years, combined with temperature data from Tamil Nadu’s official sources. We employed feature engineering techniques to prepare the data for machine learning models, specifically XGBoost regressor, lasso, and ridge regression. The time series data was then analyzed using both univariate and multivariate models, including auto regressive integrated moving average (ARIMA) and vector autoregressive (VAR) models. The results show that our models can effectively forecast ED, providing critical insights for policymakers and stakeholders involved in energy planning and resource management.
Volume: 38
Issue: 1
Page: 668-676
Publish at: 2025-04-01

Machine learning-based intelligent result compilation RPA bot for higher education institutions

10.11591/ijeecs.v38.i1.pp587-594
Neelam Yadav , Supriya P. Panda
Educators are essential for societal progress, and well-educated students are pivotal for a promising future. Higher education faces challenges such as budget constraints, limited time, and a shortage of trained personnel, leading to faculty stress. Emerging technologies such as artificial intelligence (AI), machine learning (ML), and block chain provide solutions, with robotic process automation (RPA) bots a notable advanced AI subfield-automating repetitive tasks, thereby freeing teachers to focus on more essential responsibilities. RPA bots automate various educational processes, including examinations, admissions, marks updating, student record management, result compilation, human resources, resume screening, and administration. This research examines robotic automation in higher education institutions (HEIs), selecting and prioritizing RPA tasks through a survey involving subject matter experts (SMEs) from different HEIs, including professors and RPA experts. The research aims to develop a “virtual software bot” for automating “result compilation” post-examination. Using tools like XPATH, Whisper, and the web-based automation program Selenium web in Python, the bot automates this process. The ML library “Whisper” addresses the reCAPTCHA problem. The automated bot generates comma separated values (CSV) files in specific formats, completing the task 58 times faster than humans and saving 43 man-hours by compiling results for 653 students in 45 minutes.
Volume: 38
Issue: 1
Page: 587-594
Publish at: 2025-04-01

Experimental research on text CAPTCHA of fine-grained security features

10.11591/ijeecs.v38.i1.pp535-545
Qian Wang , Shafaf Ibrahim , Xing Wan , Zainura Idrus
CAPTCHA is a cybersecurity measure that distinguishes between humans and automated scripts. Researchers have employed various security features to thwart automated program identification by hackers. However, previous research on the attack resistance of CAPTCHAs has used roughly quantitative analysis instead of a fine-grain quantitative study. This study implemented comparative experiments based on CAPTCHA recognition algorithms to find the best-mixed security features. A multi-stage best parameter selection (MBPS) mechanism was proposed in this study. Experiment results indicated that mixed security features of “overlap + scale + rotate + bg (background)” were the best, with an average machine recognition accuracy of only 4.81%. The contrast experiment result illustrated that the anti-attack ability of mixed security features was better than adding adversarial noise, with machine recognition accuracy decreased by 2.2%. Moreover, by investigating the efficacy of security feature parameters, this study provides practical guidelines for designing robust CAPTCHAs. Furthermore, this study also presents valuable insights into the security of image generation technology.
Volume: 38
Issue: 1
Page: 535-545
Publish at: 2025-04-01

Enhancing attack detection in IoT through integration of weighted emphasis formula with XGBoost

10.11591/ijeecs.v38.i1.pp641-648
Januar Al Amien , Hadhrami Ab Ghani , Nurul Izrin Md Saleh , Soni Soni
This research addresses the challenge of detecting attacks in the internet of things (IoT) environment, where minority classes often go unnoticed due to the dominance of majority classes. The primary objective is to introduce and integrate the imbalance ratio formula (IRF) into the XGBoost algorithm, aiming to provide greater emphasis on minority classes and ensure the model's focus on attack detection, particularly in binary and multiclass scenarios. Experimental validation using the IoTID20 dataset demonstrates the significant enhancement in attack detection accuracy achieved by integrating IRF into XGBoost. This enhancement contributes to the consistent improvement in distinguishing attacks from normal traffic, thereby resulting in a more reliable attack detection system in complex IoT environments. Moreover, the implementation of IRF enhances the robustness of the XGBoost model, enabling effective handling of imbalanced datasets commonly encountered in IoT security applications. This approach advances intrusion detection systems by addressing the challenge of class imbalance, leading to more accurate and efficient detection of malicious activities in IoT networks. The practical implications of these findings include the enhancement of cybersecurity measures in IoT deployments, potentially mitigating the risks associated with cyber threats in interconnected smart environments.
Volume: 38
Issue: 1
Page: 641-648
Publish at: 2025-04-01

Hybrid horned lizard optimization algorithm-aquila optimizer for DC motor

10.11591/ijai.v14.i2.pp1673-1682
Widi Aribowo , Laith Abualigah , Diego Oliva , Toufik Mzili , Aliyu Sabo
This research presents a modification of the horned lizard optimization (HLO) algorithm to optimize proportional integral derivative (PID) parameters in direct current (DC) motor control. This hybrid method is called horned lizard optimization algorithm-aquila optimizer (HLAO). The HLO algorithm models various escape tactics, including blood spraying, skin lightening or darkening, crypsis, and cellular defense systems, using mathematical techniques. HLO enhancement by modifying additional functions of aquila optimizer improves HLO performance. This research validates the performance of HLAO using performance tests on the CEC2017 benchmark function and DC motors. From the CEC2017 benchmark function simulation, it is known that HLAO's performance has promising capabilities. By simulating using 3 types of benchmark functions, HLOA has the best value. Tests on DC motors showed that the HLAO-PID method had the best integrated of time-weighted squared error (ITSE) value. The ITSE value of HLOA is 89.25 and 5.7143% better than PID and HLO-PID.
Volume: 14
Issue: 2
Page: 1673-1682
Publish at: 2025-04-01

Flexible hybrid graphene-based NFC tag antenna for temperature monitoring application

10.11591/ijeecs.v38.i1.pp227-242
Najwa Mohd Faudzi , Ahmad Rashidy Razali , Asrulnizam Abd Manaf , Nurul Huda Abd Rahman , Ahmad Azlan Ab Aziz , Syed Muhammad Hafiz , Suraya Sulaiman , Nora’zah Abdul Rashid , Amirudin Ibrahim , Aiza Mahyuni Mozi
A hybrid graphene-based material, composed of reduced graphene oxide (rGO) and silver nanoparticle (AgNP), has been proposed for a near field communication (NFC) tag antenna with an integrated, flexible temperature monitoring circuit. The limited availability of high-conductivity graphene-based materials in the market has restricted the use of graphene in NFC tag applications. Therefore, this paper proposes a hybrid graphene-based composition featuring a high conductivity of 3.95×106 S/m. The feasibility of this material for NFC tags had not been validated previously, which is the main motivation for this research. The synthesis of the materials, along with the design, fabrication, and characterization of the NFC tag, is also presented. Results show that the inkjet-printed tag achieves a good reading range of up to 3 cm and demonstrates robustness against bending from 60⁰ to 190⁰, maintaining a maximum reading range of 1.3 cm. Performance on various materials, such as plastic, paper, and carton, also shows minimal impact on frequency shifting. Additionally, the graphene-based NFC tag integrates well with the temperature circuit, effectively monitoring temperatures in the 20-60 ⁰C range in real-time. This makes the developed tag suitable for applications such as food safety monitoring systems through NFC-integrated packaging.
Volume: 38
Issue: 1
Page: 227-242
Publish at: 2025-04-01

Identification of ocular disease from fundus images using CNN with transfer learning

10.11591/ijeecs.v38.i1.pp613-621
Fatima Zohra Berrichi , Abderrahim Belmadani
Eye diseases are one of the serious health problems affecting human life. Detecting and diagnosing them early is critical to prompt treatment and preventing vision loss. However, all studies in the field of eye disease classification using machine learning models are limited to the detection of single diseases, and the accuracy rate is still low in multi-class systems. In this study, we propose a multi-class classification model using four pre-trained CNNs (DenseNet121, ResNet50, EfficientNetB3 and VGG16). The model classified eye diseases into four categories: diabetic retinopathy, cataract, glaucoma, and normal. To improve the training process, another data augmentation technique is applied to increase the amount of data. The performance metrics of the system are calculated using the confusion matrix. DenseNet-121 shows excellent performance in retinal disease classification in 30 epochs of training, with training and test accuracy reaching 99.97% and 96.21% respectively. The implementation of this system should be considered as a very useful means to help ophthalmologists to rapid and precision detection of various eye diseases in the future.
Volume: 38
Issue: 1
Page: 613-621
Publish at: 2025-04-01

Fake review detection using enhanced ensemble support vector machine system on e-commerce platform

10.11591/ijeecs.v38.i1.pp478-485
Seenia Joseph , S. Hemalatha
Due to the quick growth of online marketing transactions, including buying and selling, fake reviews are created to promote the product market and mislead new customers. E-commerce customers can post reviews and comments on the goods or services they obtained. Before making a purchase, new customers frequently read the feedback and comments posted on the website. Nowadays customers find it very difficult to identify whether the reviews are fake or not, but doing so is essential. So, it's very crucial to develop an online spam detection system to help both consumers and producers in their decision-making. The reviewer's behaviour and important review characteristics can help you identify fake reviews. The importance of this study is to develop a fake review detection system on e-commerce platforms using an enhanced ensemble support vector machine system in which the Euclidean distance is replaced with the Mahalanobis distance metric. Review texts collected from Amazon and Yelp were given as input data sets into the constructed model and classified as fake or real.
Volume: 38
Issue: 1
Page: 478-485
Publish at: 2025-04-01

Boosting industrial internet of things intrusion detection: leveraging machine learning and feature selection techniques

10.11591/ijai.v14.i2.pp1232-1241
Lahcen Idouglid , Said Tkatek , Khalid Elfayq
The rapid integration of industrial internet of things (IIoT) technologies into Industry 4.0 has revolutionized industrial efficiency and automation, but it has also exposed critical vulnerabilities to cyber threats. This paper delves into a comprehensive evaluation of machine learning (ML) classifiers for detecting anomalies in IIoT environments. By strategically applying feature selection techniques, we demonstrate significant enhancements in both the accuracy and efficiency of these classifiers. Our findings reveal that feature selection not only boosts detection rates but also minimizes computational demands, making it a cornerstone for developing resilient intrusion detection systems (IDS) tailored for Industry 4.0. The insights garnered from this study pave the way for deploying more robust security frameworks, safeguarding the integrity and reliability of IIoT infrastructures in modern industrial settings.
Volume: 14
Issue: 2
Page: 1232-1241
Publish at: 2025-04-01
1 2 3
...
1995 1996
Show 218 of 1996

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