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

23,598 Article Results

Photovoltaic power prediction using deep learning models: recent advances and new insights

10.11591/ijece.v14i5.pp5926-5940
Basma Saad , Asmaa El Hannani , Abdelhak Aqqal , Rahhal Errattahi
Artificial intelligence (AI) and its application across various domains have sparked significant interest, with each domain presenting distinct characteristics and challenges. In the renewable energies sector, accurate prediction of power output from photovoltaic (PV) panels using AI is crucial for meeting energy demand and facilitating energy management and storage. The field of data analysis has grown rapidly in recent years, with predictive models becoming increasingly popular for forecasting and prediction tasks. However, the accuracy and reliability of these models depend heavily on the quality of data, data preprocessing, model learning and evaluation. In this context, this paper aims to provide an in-depth review of previous research and recent progress in PV solar power forecasting and prediction by identifying and analyzing the most impacting factors. The findings of the literature review are then used to implement a benchmark for PV power prediction using deep learning models in different climates and PV panels. The aim of implementing this benchmark is to gain insights into the challenges and opportunities of PV power prediction and to improve the accuracy, reliability and explainability of predictive models in the future.
Volume: 14
Issue: 5
Page: 5926-5940
Publish at: 2024-10-01

Electronic document management systems implementation across industries: systematic analysis

10.11591/ijeecs.v36.i1.pp264-273
Dian Anggraini , Kusworo Adi , Jatmiko Endro Suseno
The construction sector’s pivotal role in the global economy faces challenges due to its dynamic nature. Inaccurate documentation impacts project cost management, underscoring the need for effective document management systems (DMS), including electronic document management systems (EDMS). This study conducts a systematic literature review to comprehensively examine EDMS implementation, utilization, and effectiveness across sectors. Analyzing peer-reviewed articles and scholarly sources reveals key themes, trends, and findings, providing insights into successful EDMS adoption and best practices. The review contributes evidence-based insights for practitioners, researchers, and policymakers, addressing gaps in knowledge and advancing understanding of EDMS in modern information management. Additionally, it presents a detailed breakdown of publication distribution across sectors, highlighting significant research areas like companies and businesses, education, and information technology and software. Furthermore, analysis of factors influencing employee behavior, including technical factors, employee’s personal characteristics, organizational factors, and trust, offers valuable insights into workplace dynamics. Overall, the study offers comprehensive insights into EDMS implementation, guiding future research, and organizational strategies.
Volume: 36
Issue: 1
Page: 264-273
Publish at: 2024-10-01

Empowering E-learning through blockchain: an inclusive and affordable tutoring solution

10.11591/ijece.v14i5.pp5554-5565
Saadia Lgarch , Meriem Hnida , Asmaa Retbi
This study presents an innovative approach using the Ethereum blockchain to democratize access to tutoring services, advancing educational technology by bridging the affordability gap for learners with limited financial resources. This solution enables low-income learners to access tutoring services without significant expenses by eliminating intermediaries through smart contracts. Learners can directly book tutoring services based on fees and evaluations, ensuring a fair and accessible experience. The findings show that this approach reduces tutoring expenses and improves trust and accountability through transparent transactions and feedback mechanisms. The proposed system demonstrates how blockchain technology can foster a more equitable and efficient educational landscape, offering personalized
Volume: 14
Issue: 5
Page: 5554-5565
Publish at: 2024-10-01

The surprising influence of social commerce service quality on purchase intentions mediated by e-commerce

10.11591/ijeecs.v36.i1.pp367-374
Bukky Suwarno , Wawan Dhewanto , Prawira Fajarindra Belgiawan
Social commerce has become a recent phenomenon and is poised to grow rapidly in the next few years. To better address customer behavior on social commerce platforms, it is imperative to acquire a comprehensive understanding of social commerce from the perspective of service quality. The objective of this research is to examine the dimensions of social commerce service quality and to reveal the factors influencing purchase intentions among the expanding user population. This study identified seven critical dimensions of social commerce service quality (website design, fulfilment, customer service, communication, contact, credibility, and security) that influence purchase intention. This research adopts a questionnaire survey method to collect data from social commerce users. Using PLS-SEM, the findings from an empirical analysis, conducted with a sample of 411 social commerce users, demonstrate that all measured dimensions significantly impact the intention to purchase. The findings also demonstrate that e-commerce has considerable influence on customer purchase intention as a partial mediator in social commerce. The findings hold significant implications for social commerce enterprises to increase customer attraction by identifying the motivations behind their purchasing decisions.
Volume: 36
Issue: 1
Page: 367-374
Publish at: 2024-10-01

Forecasting creditworthiness in credit scoring using machine learning methods

10.11591/ijece.v14i5.pp5534-5542
Ayagoz Mukhanova , Madiyar Baitemirov , Azamat Amirov , Bolat Tassuov , Valentina Makhatova , Assemgul Kaipova , Ulzhan Makhazhanova , Tleugaisha Ospanova
This article provides an overview of modern machine learning methods in the context of their active use in credit scoring, with particular attention to the following algorithms: light gradient boosting machine (LGBM) classifier, logistic regression (LR), linear discriminant analysis (LDA), decision tree (DT) classifier, gradient boosting classifier and extreme gradient boosting (XGB) classifier. Each of the methods mentioned is subject to careful analysis to evaluate their applicability and effectiveness in predicting credit risk. The article examines the advantages and limitations of each method, identifying their impact on the accuracy and reliability of borrower creditworthiness assessments. Current trends in machine learning and credit scoring are also covered, warning of challenges and discussing prospects. The analysis highlights the significant contributions of methods such as LGBM classifier, LR, LDA, DT classifier, gradient boosting classifier and XGB classifier to the development of modern credit scoring practices, highlighting their potential for improving the accuracy and reliability of borrower creditworthiness forecasts in the financial services industry. Additionally, the article discusses the importance of careful selection of machine learning models and the need to continually update methodology in light of the rapidly changing nature of the financial market.
Volume: 14
Issue: 5
Page: 5534-5542
Publish at: 2024-10-01

Enhancing the resistance of password hashing using binary randomization through logical gates

10.11591/ijece.v14i5.pp5400-5407
Muhamad Zaki Anbari , Bambang Sugiantoro
Digitalization in various sectors makes information security issues very crucial. Information security follows the authentication, authorization, and accounting (AAA) principle, where one of the most important parts is authentication. The most widely used authentication method is username-password. The best method to secure a user-pass is to convert the plaintext using a hash so that the converted plaintext cannot be recovered. However, with higher technology, hackers can crack the ciphertext using brute force. This research proposes a username-password scrambling algorithm before it is fed into the hash function to improve resilience from attacks. This algorithm is named logical gates (LG). It works by converting the user pass into binary form, adding salt, and scrambling it with certain logical gates before inserting it into the hash function. Testing is divided into two: time of execution and attack resistance. Time of execution results show that LG takes 0.0443432033 s, while without LG takes 0.01403197646 s. The resistance of attack results show that the plaintext of the hash amplified by LG cannot be cracked at all and increases the attack time by 321.3% at prefix and 161.3% at postfix, while without LG, the plain text can be found for a certain duration of time.
Volume: 14
Issue: 5
Page: 5400-5407
Publish at: 2024-10-01

Effect of Na-EDTA on electrical characteristics NaCl electrolyte battery charging solar panels

10.11591/ijece.v14i5.pp4846-4855
Dina Maizana , Moranain Mungkin , Habib Satria , Syafii Syafii , Muhammad Fadlan Siregar
This research investigates the problem of Cu-Zn electrode batteries with NaCl electrolyte. Previous studies have indicated problems with the electrolyte and electrodes after charging, such as turbidity and deposits in the electrolyte, as well as corrosion on the electrodes. Consequently, the battery can only be used once due to a decline in its electrical characteristics after the initial charging. Through this research, improvements were made to the electrical characteristics of the battery by adding Na-EDTA to enhance usage efficiency. The research method involved mixing NaCl solution with the highest electrical conductivity, using six pairs of Cu-Zn electrodes arranged in series. The physical conditions of the electrolyte and electrodes were observed, and electrical characteristics were measured. The research results indicate that the use of NaCl+Na-EDTA electrolyte produces a battery voltage of 4.20 volts with a current of 2 Ah and can be used twice. Charging with solar panels can be done in 1 hour, but the frequency is limited to two times.
Volume: 14
Issue: 5
Page: 4846-4855
Publish at: 2024-10-01

Empowering crop cultivation: harnessing internet of things for smart agriculture monitoring

10.11591/ijece.v14i5.pp6023-6035
Jamil Abedalrahim Jamil Alsayaydeh , Mohd Faizal Yusof , Mithilanandini S. Magenthiran , Rostam Affendi Hamzah , Izadora Mustaffa , Safarudin Gazali Herawan
Agriculture, the foundation of human civilization, has relied on manual practices in the face of unpredictable weather for millennia. The contemporary era, however, witnesses the transformative potential of the Internet of things (IoT) in agriculture. This paper introduces an innovative IoT-driven smart agriculture system empowered by Arduino technology, making a significant contribution to the field. It integrates key components: a temperature sensor, a soil moisture sensor, a light-dependent resistor, a water pump, and a Wi-Fi module. The system vigilantly monitors vital environmental parameters: temperature, light intensity, and soil moisture levels. Upon surpassing 30°C, an automatic cooling fan alleviates heat stress, while sub-300CD light levels trigger light-emitting diode lighting for optimal growth. Real-time soil moisture data is relayed to the “Blynk” mobile app. Temperature thresholds align with specific crops, and users can manage the water pump via Blynk when manual intervention is required. This work advances agricultural practices, optimizing water management by crop type. Through precise coordination of soil moisture, temperature, and light intensity, the system enhances productivity while conserving water resources and maintaining fertilizer balance.
Volume: 14
Issue: 5
Page: 6023-6035
Publish at: 2024-10-01

Utilizing digital elevation models and geographic information systems for hydrological analysis and fire prevention in Khuan Kreng peat swamp forest, Southern Thailand

10.11591/ijece.v14i5.pp5408-5419
Uraiwun Wanthong , Somporn Ruang-On , Nunticha Limchoowong , Phitchan Sricharoen , Panjit Musik
The objectives of this research were to create a topographic model using Mathematica and hydrologic model using ArcGIS for water management aimed at preventing forest fires in the Khuan Kreng peat swamp forest. Pan basin area in Kreng Sub-district, characterized by low mountains, where the Cha-Uat canal intersects the krajood forest, was revealed by the hydrographic model. Kreng Sub-district was traversed by three main streams: Khuan canal, Hua Pluak Chang canal, and Laem canal. Additionally, several tributary canals that interconnect, ultimately converging into the Cha-Uat Phraek Muang canal were identified. During the dry period, the water from these canals flowed into the Cha-Uat Phraek Muang canal. To mitigate the risk of fires, it was essential to install water table measuring devices and underground barrier gates at the drain points. This ensured the return of water from the Cha-Uat Phraek Muang canal to the Khuan Kreng peat swamp forest. Maintaining sufficient water table level was crucial, as the occurrence of fires was more likely when the water table dropped below the soil surface. When the swamp forest was adequately hydrated, wildfires were confined to a narrow area since they could only burn on the forest surface, which was easier to extinguish.
Volume: 14
Issue: 5
Page: 5408-5419
Publish at: 2024-10-01

Histopathological cancer detection based on deep learning and stain images

10.11591/ijeecs.v36.i1.pp214-230
Dina M. Ibrahim , Mohammad Ali A. Hammoudeh , Tahani M. Allam
Colorectal cancer (CRC)-a malignant growth in the colon or rectum- is the second largest cause of cancer deaths worldwide. Early detection may increase therapy choices. Deep learning can improve early medical detection to reduce the risk of unintentional death from an incorrect clinical diagnosis. Histopathological examination of colon cancer is essential in medical research. This paper proposes a deep learning-based colon cancer detection method using stain-normalized images. We use deep learning methods to improve detection accuracy and efficiency. Our solution normalizes image stain variations and uses deep learning models for reliable classification. This research improves colon cancer histopathology analysis, which may enhance diagnosis. Our paper uses DenseNet-121, VGG-16, GoogLeNet, ResNet-50, and ResNet-18 deep learning models. We also analyze how stain normalization (SN) improves our model on histopathology images. The ResNet-50 model with SN yields the highest values (9.94%) compared to the other four models and the nine models from previous studies.
Volume: 36
Issue: 1
Page: 214-230
Publish at: 2024-10-01

Rotor angle deviation regulator to enhance the rotor angle stability of synchronous generators

10.11591/ijece.v14i5.pp4879-4887
Nor Syaza Farhana Mohamad Murad , Muhammad Nizam Kamarudin , Muhammad Iqbal Zakaria
Occurrences of disturbance affect the rotor angle operation of a synchronous generator in the generation system of a power system. The disturbance will disrupt the synchronous generator's rotor oscillation and result in rotor angle instability, which will degrade the power system's performance. This paper aims to develop a Lyapunov-based rotor angle deviation regulator for the nonlinear swing equation of a synchronous generator. The proposed regulator is expected to assure asymptotic stability of the rotor angle and robustness to uncertainty. Backstepping and Lyapunov redesign techniques are employed in developing the regulator. To validate the effectiveness and robustness of the regulator, a simulation in MATLAB/Simulink is carried out. The simulation result shows that the asymptotic stability and robustness of the regulator are guaranteed regardless of the disturbance.
Volume: 14
Issue: 5
Page: 4879-4887
Publish at: 2024-10-01

Performance evaluation of single-mode fiber optic-based surface plasmon resonance sensor on material and geometrical parameters

10.11591/ijece.v14i5.pp5072-5082
Imam Tazi , Dedi Riana , Mohamad Syahadi , Muthmainnah Muthmainnah , Wiwis Sasmitaninghidayah , Lia Aprilia , Wildan Panji Tresna
Surface plasmon resonance (SPR) sensors are proficient at detecting minute changes in refractive index, making them ideal for biomolecule detection. Traditional prism-based SPR sensors encounter miniaturization challenges, encouraging exploration of alternatives like fiber optic-based SPR (FO-SPR) sensors. This study comprehensively investigates the effects of material and geometrical parameters on the performance of single-mode FO-SPR sensors using Maxwell's equation solver software based on the finite-difference time-domain (FDTD) method. The findings highlight the influence of plasmonic thin film materials and thickness on SPR spectrum profiles and sensitivity. Silver (Ag) demonstrates superior performance compared to copper (Cu) and gold (Au) in transmission type, achieving a sensitivity of up to 2×103 nm/RIU, while the sensitivities of Cu and Au are lower. Probe length and core diameter impact spectrum profiles, specifically resonance depth, without affecting sensitivity. Furthermore, variations in core refractive index influence both spectrum profiles and sensitivity. Probe types significantly affect both spectrum profiles and sensitivity, with the reflection type surpassing the transmission type. These results provide suggestions for optimizing FO-SPR sensors in biotechnological applications.
Volume: 14
Issue: 5
Page: 5072-5082
Publish at: 2024-10-01

Optimizing intrusion detection in 5G networks using dimensionality reduction techniques

10.11591/ijece.v14i5.pp5652-5671
Zaher Salah , Esraa Elsoud , Waleed Al-Sit , Esraa Alhenawi , Fuad Alshraiedeh , Nawaf Alshdaifat
The proliferation of internet of things (IoT) technologies has expanded the user base of the internet, but it has also exposed users to increased cyber threats. Intrusion detection systems (IDSs) play a vital role in safeguarding against cybercrimes by enabling early threat response. This research uniquely centers on the critical dimensionality aspects of wireless datasets. This study focuses on the intricate interplay between feature dimensionality and intrusion detection systems. We rely on the renowned IEEE 802.11 security-oriented AWID3 dataset to implement our experiments since AWID was the first dataset created from wireless network traffic and has been developed into AWID3 by capturing and studying traces of a wide variety of attacks sent into the IEEE 802.1X extensible authentication protocol (EAP) environment. This research unfolds in three distinct phases, each strategically designed to enhance the efficacy of our framework, using multi-nominal class, multi-numeric class, and binary class. The best accuracy achieved was 99% in the three phases, while the lowest accuracy was 89.1%, 60%, and 86.7% for the three phases consecutively. These results offer a comprehensive understanding of the intricate relationship between wireless dataset dimensionality and intrusion detection effectiveness.
Volume: 14
Issue: 5
Page: 5652-5671
Publish at: 2024-10-01

Dual soft decoding of linear block codes using memetic algorithm

10.11591/ijece.v14i5.pp5263-5273
Rajaa Sliman , Ahmed Azouaoui
In this article we will approach the soft-decision decoding for the linear block codes, is a kind of decoding algorithms used to decode data to form better original estimated received message, it is considered as a NP-hard problem. In this article we present a new decoder using memetic algorithm such metaheuristic technic operates on the dual code rather than the code itself that aims to find the error caused when sending a codeword calculated from a message of k bits of information, the resulting codeword contains n bits, including the redundancy bits, the efficiency of an error-correcting code is equivalent to the ratio k/n, the rate is belong the interval [0,1]. Hence a good code is the one that ensures a certain error correcting capability at minimum ratio. The results proved that this approach using a combination of genetic algorithm and local search algorithm provides a sufficiently good solution to an optimization problem; the new decoder is applied on linear codes where the structure is given by a parity check matrix.
Volume: 14
Issue: 5
Page: 5263-5273
Publish at: 2024-10-01

Optimal control design of the COVID-19 model based on Lyapunov function and genetic algorithm

10.11591/ijece.v14i5.pp5117-5130
Aminatus Sa'adah , Roberd Saragih , Dewi Handayani
Millions of people died worldwide as a result of the coronavirus disease 2019 (COVID-19) pandemic that started in early 2020. Examining the COVID-19 susceptible-exposed-infected-recovery (SEIR) mathematical model is one approach to developing the best control scenario for this disease. The study utilized two control variables, vaccination, and therapy, to construct a control function that relied on the quadratic Lyapunov function. The control objective was to lower the number of COVID-19 infections while maintaining system stability. A genetic algorithm (GA) is used as a novel method to estimate controller parameter value to replace the previously used parameter tuning procedure. Then, a numerical simulation was carried out implementing three control scenarios, namely vaccination control only, treatment control only, and vaccination and treatment control simultaneously. Based on the results, scenario 3 (vaccination and treatment simultaneously) showed the most significant decrease: the average decrease in the exposed human population was 98.29%, and the infected human population was 98.18%.
Volume: 14
Issue: 5
Page: 5117-5130
Publish at: 2024-10-01
1 2 3
...
1573 1574
Show 15 of 1574

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