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

29,734 Article Results

Improving multilabel classification of hate speech and abusive language in Indonesian using MAML

10.12928/telkomnika.v24i2.27332
Jasman; Institut Teknologi Nasional Bandung Pardede , Ghixandra; Institut Teknologi Nasional Bandung Julyaneu Irawadi , Rizka; Institut Teknologi Nasional Bandung Milandga Milenio
This study investigates automated multi-label detection of hate speech and abusive language (HSAL) in Indonesian social media, addressing challenges of data imbalance, especially in minority labels. Two training approaches are compared: standard supervised learning and meta-learning using the model-agnostic meta-learning (MAML) algorithm. IndoBERTweet-BiGRU is adopted as the baseline model, while MAML is leveraged to enhance generalization and adaptability with limited training data. Both models are trained on a multilabel dataset with 13 HSAL categories exhibiting highly imbalanced distributions. The best supervised model achieved an F1-Micro of 84.02% and an F1-macro of 77.97%, whereas the best MAML-trained model reached 84.12% and 76.85%, respectively. Although the overall gap is small, MAML demonstrates notable improvements on minority classes such as hate speech (HS) physical, gender, and race, shown through higher F1-score and area under the receiver operating characteristic curve (AUROC) values. These results highlight its strength in low-resource classification settings. This study is limited to Indonesian language and YouTube transcript contexts, and MAML incurs higher training complexity. Cultural and linguistic nuances also present potential bias in real-world use. Despite these constraints, the proposed system offers practical benefits by enabling fine-grained HSAL classification and supporting earlier detection of harmful online content.
Volume: 24
Issue: 2
Page: 549-563
Publish at: 2026-04-01

Attributes conducive to anthropomorphism in artificial intelligence

10.12928/telkomnika.v24i2.27483
Rizwan; Murray State University Syed , Hassan; Murray State University Mistareehi
The rapid development of artificial intelligence (AI), particularly large language models (LLMs), has generated both enthusiasm and concern regarding its role in society. While these systems demonstrate impressive technical capabilities, public acceptance is often hindered by perceptions of unpredictability, mistrust, and fears amplified by media narratives. One potential strategy to improve user acceptance is anthropomorphism, the attribution of human-like qualities to AI systems which can make interactions feel more natural and trustworthy. This paper investigates the attributes most conducive to anthropomorphism by conducting a structured review across psychology, human-robot interaction, communication studies, and business applications. The analysis identifies key traits such as emotional expressiveness, conversational coherence, adaptive social behavior, and role-based framing that enhance perceptions of AI as relatable and dependable. By synthesizing these insights, we propose a conceptual framework that highlights the psychological, social, and technical dimensions of anthropomorphism in AI. The findings provide guidance for designing AI systems that balance efficiency with user trust, thereby supporting more effective integration of AI into business, research, and everyday life.
Volume: 24
Issue: 2
Page: 588-598
Publish at: 2026-04-01

Business intelligence for measuring global systems for mobile communication provider performance

10.12928/telkomnika.v24i2.26293
Yusri Eli Hotman; Universitas Trisakti Turnip , Dedy; Trisakti University Sugiarto , Rina; Universitas Trisakti Fitriana , Yun-Chia; Yuan Ze University Liang
Internet access is getting easier in various places, including Indonesia. Telecommunication media are no longer dominated by the use of pulse signals but have shifted to relying on internet access. This study aims to create a data visualization of internet speed in Bekasi urban sub-districts using the business intelligence (BI) model with online analytical processing (OLAP). Clustering was carried out using two methods, namely the K-means and K-medoids methods which were selected based on the Davies Bouldin index (DBI) value. This study produced a visual data prototype from the results of clustering from the data mining process and was accompanied by supporting data in the form of information on the highest and lowest speeds in the studied sub-districts. The clustering process uses K-means for uploading data with a DBI value of 0.847, while the data download uses K-medoids with a DBI value of 0.871. The prototype displays observation data, maximum and minimum value information, and the clustering result. The functional test result for the prototype showed conformity with the requirements, while the validation test showed that the prototype passed the validation test with a score of 0.8833.
Volume: 24
Issue: 2
Page: 737-750
Publish at: 2026-04-01

An interpretable deep learning framework for early detection of depression using hybrid architectures

10.11591/ijece.v16i2.pp895-904
Chaithra Indavara Venkateshagowda , Roopashree Hejjajji Ranganathasharma , Yogeesh Ambalagere Chandrashekaraiah
Current techniques for detecting depression are labor-intensive and subjective, depending on clinical interviews or self-reports. There is a growing adoption of machine learning (ML) and natural language processing (NLP) to automatically identify depression in textual data. The lack of interpretability, which is essential for healthcare applications, is still a major obstacle, though. By combining convolution neural network (CNN) for feature extraction, bidirectional long short-term memory (BiLSTM) for capturing sequential dependencies, and transformer-based pre-trained language model (PTLM) for contextual understanding, this study offers an interpretable framework for early depression identification. Additionally, the system uses a novel interpretability method to guarantee transparent decision-making. The outcome of the proposed system is found to achieve 96.2% accuracy, 94.5% precision, 95.1% recall, and 94.8% F1-score, which is a significant improvement over current method. This framework acts as a useful tool for early mental health intervention.
Volume: 16
Issue: 2
Page: 895-904
Publish at: 2026-04-01

Identification of critical buses in the Sulbagsel electrical system network integrated with wind power plants

10.11591/ijece.v16i2.pp587-597
Andi Muhammad Ilyas , Agus Siswanto , Muhammad Natsir Rahman
The growing deployment of renewable energy has become increasingly important as conventional fossil-based generation faces sustainability and resource limitations. On Sulawesi Island, Indonesia, wind energy contributes to the regional grid through several wind power plants, whose fluctuating generation introduces operational concerns for system stability. This study investigates the stability performance of the Sulbagsel 78-bus network by pinpointing vulnerable buses and examining the effects of wind power variability. A hybrid stability index (HSI), which integrates multiple stability indicators, is applied to obtain a more robust assessment. The analysis shows that the entire system operates within a secure margin, with all index values remaining below the critical limit (<1). The most sensitive areas are located on the transmission paths connecting Bus 56 Sidera–Bus 57 Sidera 70 kV (0.02268), Bus 38 Bosowa–Bus 40 Pangkep (0.02220), and Bus 73 Powatu 150 kV–Bus 74 Powatu 70 kV (0.02187). In contrast, the Bus 24 Tanjung Bunga–Bus 25 Bontoala corridor demonstrates the strongest stability margin (0.00026). These results indicate that the variability of wind generation does not impose significant negative impacts on the overall stability of the Sulbagsel power system.
Volume: 16
Issue: 2
Page: 587-597
Publish at: 2026-04-01

Elitist genetic algorithm improved with parenting fitness parameter

10.11591/ijece.v16i2.pp883-894
Ouiss Mustapha , Ettaoufik Abdelaziz , Marzak Abdelaziz
In genetic algorithms, the selection of individuals that will be part of future generations is a critical process of the algorithm. Various strategies exist to select these individuals: the general approach and the elitist approach. The general approach involves replacing the whole current population with the offspring generated so far. The elitist approach introduces a competitive element in which both parents and offspring compete for survival, and only fit individuals will be part of the next generation. While selecting fit individuals helps the algorithm to produce better results, the elitism has a major drawback: the premature convergence, which can limit the algorithm's overall performance. In this article, we compared a typical elitist genetic algorithm and an elitist algorithm improved with the parenting fitness parameter in resolving the vehicle routing problem with drones (VRPD). The parenting fitness parameter helps preserving diversity by retaining parents with high offspring potential despite of their personal fitness. The findings from the study demonstrates that integrating the parenting fitness parameter lead to better results in comparison with a typical elitist genetic algorithm, with relative improvement varying from 1.06% to 10.34% according to the dataset’s size.
Volume: 16
Issue: 2
Page: 883-894
Publish at: 2026-04-01

Hyperparameter tuning of MobileNetV2 on forest and land fire severity classification

10.11591/ijece.v16i2.pp964-972
Assad Hidayat , Imas Sukaesih Sitanggang , Lailan Syaufina
Forest and land fires pose significant environmental challenges, causing economic and ecological damage depending on their severity. This study proposes a deep learning-based classification model to assess fire severity using the MobileNetV2 architecture. A dataset of 560 post-fire images was categorized into five severity levels, with dataset preprocessing involving resizing, rescaling, and image augmentation. To enhance model performance, K-means clustering was applied for balanced data distribution across classes. The model was trained using grid search for hyperparameter tuning, with the optimal combination being a batch size of 8, learning rate of 0.0001, and dropout of 0.3. Training was conducted in 50 epochs, and evaluation using the confusion matrix demonstrated an accuracy of 85%, precision of 86%, and recall of 81%. The results indicate that MobileNetV2 effectively classifies post-fire severity levels, offering a reliable tool for post-disaster assessment. This study highlights the significance of dataset preprocessing and hyperparameter tuning in improving model accuracy. Future research should explore alternative architectures and expand the dataset to enhance model generalization. These findings can aid authorities in assessing fire impact, supporting mitigation strategies, and improving post-fire land management.
Volume: 16
Issue: 2
Page: 964-972
Publish at: 2026-04-01

Single-stage single-phase grid connected inverter proportional resonant and maximum power point tracking controllers for enhanced photovoltaic system performance

10.11591/ijece.v16i2.pp651-662
Abdelaziz Kabba , Abdellah Lassioui , Hassan El Fadil
The paper develops a current control methodology for a single-phase grid-tied DC/AC inverter applied to photovoltaic (PV) energy conversion systems. It incorporates an algorithm for finding the optimal voltage and current points to obtain maximum power point tracking (MPPT), the purpose of which is to ensure better energy extraction. This is followed by a proportional-integral (PI) controller to generate the reference current. In addition, a proportional-resonant (PR) controller is used to infinitely amplify the fundamental frequency signal, which makes it possible to eliminate the steady-state error. The analytical foundations of the PR controller are presented and substantiated through simulation studies implemented in MATLAB/Simulink. The phase-locked loop (PLL) is used for synchronization, enabling accurate phase detection of the grid voltage for effective power injection. An LCL filter is also implemented between the inverter and the grid. The results provided by the dedicated software confirm the effectiveness of the proposed control system.
Volume: 16
Issue: 2
Page: 651-662
Publish at: 2026-04-01

Internet of things-based smart control and comfort classification system for broiler chicken coops using k-nearest neighbor algorithm

10.11591/ijece.v16i2.pp1039-1050
Khodijah Amiroh , Helmy Widyantara , Muhammad Dwi Hariyanto
The poultry industry increasingly relies on environmental automation to improve broiler chicken welfare and productivity. Prior studies have implemented threshold-based systems to control coop conditions, typically activating actuators based on fixed values of temperature or humidity. However, such systems lack adaptability to dynamic environmental interactions and often result in inefficient energy use and overactivation. This study proposes a novel low-cost internet of things (IoT)-based smart poultry coop system that combines real-time environmental sensing with comfort classification using the k-nearest neighbor (KNN) algorithm. The system monitors temperature, humidity, and ammonia levels through affordable sensors integrated with an ESP32 microcontroller, then transmits data via message queuing telemetry transport (MQTT) to a remote server for classification and control decision-making. Control logic is applied to activate fans, heating lamps, or humidifiers accordingly. Evaluation on a mini coop prototype demonstrated a classification accuracy of 92.2% and a 34% reduction in actuator overactivation compared to threshold-based systems. Environmental stability improved by 23%, and energy usage decreased by 12.6%. The system also features user interfaces via Telegram and Blynk, proven intuitive through informal testing. These results validate the feasibility of integrating machine learning into small-scale poultry environments, offering an intelligent, scalable, and user-friendly solution that outperforms traditional methods.
Volume: 16
Issue: 2
Page: 1039-1050
Publish at: 2026-04-01

Enhancing ride-hailing adoption: understanding factors influencing ride-hailing user attitudes and reuse intention

10.11591/ijece.v16i2.pp905-913
Mudjahidin Mudjahidin , Rafid Ikbar Athallah
Ride-hailing applications (RHA) have emerged as a revolutionary force in the transportation landscape, offering convenient and on-demand mobility solutions, thus gaining widespread popularity in the transportation sector. However, concerns arise as many RHA startups find it difficult to survive in Indonesia, and even big RHA startups are still at risk. RHA must preserve user reuse intent in order to ensure service continuation. Based on the innovation diffusion theory (IDT), the unified theory of acceptance and use of technology (UTAUT), and additional factors, this study examines 11 variables and their impact on consumer attitudes and reuse intention in a model of ride-hailing service adoption. An online survey was utilized to gather data from various demographic backgrounds, and managed to gather data from 240 respondents. Analysis was conducted using partial least squares structural equation modeling (PLS-SEM) to assess the correlations between the variables. The findings revealed that perceived usefulness, perceived ease of use, perceived risk, compatibility, and personal innovation significantly influenced consumer attitudes. Additionally, it was shown that the attitude variable and customer reuse intention were positively and significantly correlated. Based on this outcome, recommendations were made to RHA providers to improve user attitudes and intentions to reuse.
Volume: 16
Issue: 2
Page: 905-913
Publish at: 2026-04-01

Architectural trade-offs: comparative analysis across K3s, serverless, and traditional server deployments

10.11591/ijece.v16i2.pp873-882
Prajwal P. , Naveen B. Teli , Nishal H. N. , Nimisha Dey , Pratiba Deenadhayalan , Ramakanth Kumar Pattar , Pavithra Hadagali , Skanda P. R.
In modern software architecture, combining serverless computing, microservices, and containers improves scalability, performance, observability, and resilience. However, choosing the right deployment strategy is crucial. Current individual deployment methods often limit productivity because of poor integration options. This study looks at three deployment approaches: Kubernetes cluster, AWS Lambda (serverless), and Traditional Java Server. We tested performance under different workloads using virtual machines and simulations. The results show that the K3s cluster provides high throughput and low latency because it manages resources directly. AWS Lambda’s pay-as-you-go model, along with its built-in cost optimization, works well for event-driven workloads. In contrast, Java Microservice is cost-effective but needs manual tuning to control latency and error rates. Bringing these scenarios together into a single service mesh architecture could help optimize costs, performance, and system resilience.
Volume: 16
Issue: 2
Page: 873-882
Publish at: 2026-04-01

Transforming e-government projects by developing a RAF using Scrum integrated with CASE tool in Botswana

10.12928/telkomnika.v24i2.27431
Thapelo; North-West University Monageng , Bukohwo Michael; North-West University Esiefarienrhe
The digital transformation in Botswana has placed strong emphasis on e-government initiatives aimed at improving public service delivery. However, these projects continue to face low success rates due to challenges such as inadequate and reactive risk management practices, limited technical expertise, and fragmented implementation. This study proposes an integrated risk assessment framework (RAF) that combines Scrum methodology with computer-aided software engineering (CASE) tools that allows for the development of an automated, proactive, and iterative approach to risk management that is specific to the socioeconomic circumstance of Botswana. A quantitative survey was conducted with 32 project management specialists involved in e-government projects to assess their familiarity with agile methods and CASE tools, perceptions of traditional risk management approaches, and acceptance of the proposed model. The results revealed that 90.6% of respondents were familiar with Scrum, 78.1% had used CASE tools, and 81.25% supported the new framework, highlighting the urgent need for real-time risk tracking and continuous stakeholder engagement. The proposed e-government risk assessment framework (e-GRAF) model offers a flexible and adaptive solution to strengthen risk management processes, increase the success rate of e-government projects, and improve the quality and resilience of digital governance systems in Botswana.
Volume: 24
Issue: 2
Page: 466-480
Publish at: 2026-04-01

A comprehensive analysis of feature selection and XAI for machine learning classifiers to recognize guava disease

10.12928/telkomnika.v24i2.27599
Sujon Chandra; University of Frontier Technology, Bangladesh (UFTB) Sutradhar , Md. Mehedi; University of Frontier Technology Hasan
Recognizing and classifying diseases in guava is crucial for managing farms to keep crops healthy and increase harvest quality. Cultivators face the most severe challenges when it comes to recognizing and diagnosing guava fruit and leaf illnesses, a task that is nearly impossible to perform manually. This research focuses on developing a robust disease identification model using image data collected locally from guava trees. After data collection, various image processing techniques, including scaling and contrast enhancement, are utilized to make the data more suitable for use. K-means clustering is employed to quickly divide the images into groups, followed by the extraction of important characteristics. Two separate feature ranking approaches, analysis of variance (ANOVA) and least absolute shrinkage selection operator (LASSO), are used to select the best characteristics, identifying the 10 most important attributes. The adaptive boosting (AdaBoost) classifier achieves the highest accuracy among six classifiers for the top seven characteristics indicated by LASSO among the specified features. To enhance the model’s interpretability, two explanation methods, local interpretable model-agnostic explanations (LIME) and shapley additive explanations (SHAP), are employed to illustrate how the classifier reaches its conclusions. This approach not only simplifies disease identification but also clarifies the reasoning behind predictions, opening the door to real-world applications in detecting and preventing dangerous diseases.
Volume: 24
Issue: 2
Page: 574-587
Publish at: 2026-04-01

Secure two-way relaying with successive interference cancellation and fountain codes: performance analysis

10.12928/telkomnika.v24i2.27314
Nguyen Thi; Industrial University of Ho Chi Minh City Hau , Tran Trung; Posts and Telecommunications Institute of Technology Duy
This paper proposes a secure two-way relaying (TWR) scheme using fountain codes (FCs), successive interference cancellation (SIC), and digital network coding (DNC). Using FCs, two sources exchange their data by first encoding the data into a series of packets (called encoded packets). These encoded packets are then exchanged between the sources via the help of a common relay, and they are also overheard by an eavesdropper. The packet exchange is carried out over two time slots: i) in the first time slot, both sources send their encoded packets to the rela y; and ii) the relay applies SIC to decode two received packets, and then broadcasts the exclusive OR (XORed) packet to both sources in the second time slot. The sources and the eavesdropper try to collect a sufficient number of encoded packets to successfully recover the original data. This paper derives and validates exact closed-form expressions for system throughput (TP), system outage probability (SOP), and system intercept probability (SIP) over Rayleigh fading channels. Furthermore, our findings reveal a reliability-security trade-off as well as the impact of system parameters on the network performance.
Volume: 24
Issue: 2
Page: 420-430
Publish at: 2026-04-01

System dynamics control simulation for sustainability of Indonesia’s cocoa supply chain

10.12928/telkomnika.v24i2.27509
Imam; Universitas Brawijaya Santoso , Dodyk; Universitas Brawijaya Pranowo , Hendrix Yulis; Universitas Brawijaya Setyawan , Izzum; Universitas Brawijaya Wafi'uddin , Naila Maulidina; Universitas Brawijaya Lu'ayya , Annisa'u; Politeknik Negeri Jember Choirun
Indonesia’s cocoa sector faces challenges in greenhouse gas emissions and smallholder income volatility. This study develops a system dynamics model to simulate the interrelationship between carbon emissions and economic performance across the cocoa value chain, identify leverage points, and evaluate alternative policy scenarios. The model integrates environmental and economic variables into dynamic feedback structures, enabling scenario-based assessment of intervention strategies. Five scenarios were simulated: composting cocoa waste increased farmer income by 2% and reduced farm-level emissions from 0.43 to 0.303 kg CO₂-eq/kg (29.79% total reduction); biogas conversion raised income by 13.56% and reduced emissions by 11%; converting cocoa waste into animal feed slightly increased income by 0.23% while cutting emissions by 58.6%; combining composting with improved transport efficiency reduced emissions by 14%; and integrating composting, logistics optimization, and government-supported input subsidies yielded the highest performance, with a 13.50% income increase and a 70% emission reduction. These results demonstrate that integrated, system-based interventions can enhance both economic resilience and environmental sustainability. The system dynamics model provides policymakers and supply chain actors with actionable insights for designing effective, climate-aligned strategies in Indonesia’s cocoa industry.
Volume: 24
Issue: 2
Page: 431-451
Publish at: 2026-04-01
1 2 3
...
1982 1983
Show 3 of 1983

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