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

30,895 Article Results

Sentiment analysis in telecommunications: a systematic review of applications and challenges

10.12928/telkomnika.v24i4.27762
Achraf; LaGeS Laboratory, Hassania School of Public Works Bouhamidi , Naziha; LaGeS Laboratory, Hassania School of Public Works Laaz , Zineb; LaGeS Laboratory, Hassania School of Public Works Rachik
The increasing amount of digital content on online platforms creates both opportunities and challenges for telecommunications operators aiming to improve customer experience. Sentiment analysis (SA), a subfield of natural language processing (NLP), allows for the extraction of subjective information from data generated by users. This paper presents a systematic literature review (SLR), conducted in accordance with Kitchenham’s guidelines, examining the applications and challenges of SA within the telecommunications sector. A structured methodology was used to select and analyze over 100 studies published between 2017 and 2025. The review categorizes SA methods and evaluates their application in practical telecom contexts, including customer satisfaction assessment, churn prediction, service monitoring, and reputation management. Findings indicate that machine learning and deep learning models can achieve performance levels up to 97% in specific experimental contexts. However, such results may vary depending on the datasets, evaluation protocols, and application contexts. Key challenges identified include handling informal language, domain dependency, multilingualism, sarcasm detection. Finally, the review highlights future directions such as real-time sentiment tracking, multimodal analysis, and the integration of federated learning for privacy-preserving customer analytics. This review provides a foundational reference for researchers and practitioners aiming to deploy effective sentiment-driven systems in the telecom industry.
Volume: 24
Issue: 4
Page: 1224-1240
Publish at: 2026-08-01

From climate time series to planting windows in chili (Capsicum frutescens): a SARIMA–SVM–XGBoost framework with balanced-accuracy thresholding

10.11591/ijece.v16i4.pp2210-2219
Efrans Christian , Nova Noor Kamala Sari , Ressa Priskila , Septian Geges
This study proposes a spatio-temporal decision-support framework that integrates Seasonal Autoregressive Integrated Moving Average (SARIMA), Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost) to derive adaptive planting windows for chili (Capsicum frutescens) at the sub-district level. The framework addresses key challenges in climate-sensitive agriculture, including spatial data leakage and class imbalance, by employing Leave-One-Group-Out (LOGO) cross-validation and Balanced Accuracy–based threshold optimization. The proposed system transforms heterogeneous environmental data into actionable recommendations by combining climate forecasting, land suitability assessment, and yield prediction within a unified pipeline. Experimental results indicate that the framework effectively captures seasonal climate dynamics and produces consistent planting recommendations aligned with agronomic conditions, enabling multiple planting cycles per year. The primary contribution of this work lies in a transparent and generalizable integration of statistical and machine learning models into a practical decision-support framework. The proposed approach bridges predictive modeling and real-world agricultural decision-making and can be extended to other crops and regions for climate-adaptive agricultural planning.
Volume: 16
Issue: 4
Page: 2210-2219
Publish at: 2026-08-01

Indirect adaptive neural network control for constant power conversion in wave energy system

10.11591/ijece.v16i4.pp2120-2133
Jesus de la Cruz-Alejo , Hugo Beatriz Cuellar , J. Antonio Lobato Cadena , Edwin Christian Becerra-Alvarez
The conversion of ocean wave energy into electrical energy occurs near beaches and is important for the design and implementation of wave energy conversion (WEC) systems. However, its generation depends on environmental conditions, which complicates the design and control of the devices. This work presents an approach to indirect adaptive control based on artificial neural networks to detect wave conditions for the proper functioning of WEC structures. The method involves generating a constant output voltage using a voltage boost converter and a direct current-alternating current (DC-AC) converter. Maintaining a constant output power despite variations in wave conditions to generate a voltage of 24 V with a current of 2 A is the primary proposal for the control design. The mechanical design integrates a rack and pinion system and a pulley transmission that connects a floating device to an electric generator. The implementation of control is carried out on an Arduino platform. The control system was implemented on an Arduino platform, occupying 48% of the available memory, with a convergence time of 4.29 ms, a mean squared error (MSE) of 0.13715, and a root mean squared error (RMSE) of 0.37034. These low values indicate that the proposed control system has greater accuracy. The experimental results validate the proposed control system, which reduces energy conversion errors and achieves greater efficiency.
Volume: 16
Issue: 4
Page: 2120-2133
Publish at: 2026-08-01

Study of the effect of changes in cell output temperature on the net efficiency of solid oxide fuel cell power generation

10.12928/telkomnika.v24i4.27828
Handrea; PT PLN (Persero) Bernando Tambunan , Ignatius; Politeknik Negeri Bandung Riyadi Mardiyanto , Lina; Politeknik Negeri Bandung Troskialina , Sri; Politeknik Negeri Bandung Paryanto Mursid , Lidya; Politeknik Negeri Bandung Elizabeth , Dhyna; Politeknik Negeri Bandung Analyes Trirahayu , Retno; Politeknik Negeri Bandung Dwijayanti
Solid oxide fuel cell (SOFC) power plants offer a promising pathway toward highly efficient energy conversion compared to conventional systems. This study explores the impact of output temperature variations on system performance through simulation analysis. The results demonstrate that increasing the temperature difference between the fixed input (700 °C) and the cell output significantly enhances net efficiency, while gross efficiency remains relatively stable. At an output temperature of 875 °C, the system achieves a net efficiency approaching 50% and a gross efficiency of approximately 61%. These findings emphasize the critical role of output temperature in determining overall system efficiency and highlight the importance of thermal management in SOFC operation.
Volume: 24
Issue: 4
Page: 1419-1426
Publish at: 2026-08-01

Methods of finding the maximum common transitive subgraph: experimental comparison

10.12928/telkomnika.v24i4.27683
Oleg; Volgograd State Technical University Sychev , Anton; Volgograd State Technical University Chupinin
The problem of finding a maximum common subgraph (MCS) in a graph has broad applications in practical domains. However, certain scenarios require subgraphs with special properties, such as transitivity, that must be kept during building the subgraph. We formally define the concept of a transitive subgraph, investigate its properties. We study four different algorithms for finding the max imum common transitive subgraph (MCTS), compiled a list of tests aim at com paring graphs after making various changes and evaluated their accuracy and efficiency on a set of test cases. Benchmarking on 64 tests ranks the algorithms by scalability and accuracy: branch matching is the most scalable (> 1000 ver tices) and accurate (F1: 0.9907). MCS tree search is viable for graphs of up to ∼ 250 vertices (F1: 0.9752). Backtracking is limited to < 30 vertices (ac curacy: 0.5625), and brute-force is only feasible for graphs with ≤ 10 vertices, despite its high accuracy (0.9375). We discuss the advantages and disadvantages of each method, the test cases where each method demonstrates a non-optimal MCTS,identify the classes on which the methods work correctly and found that the branch matching method based on the longest common subsequence (LCS) algorithm performed the best.
Volume: 24
Issue: 4
Page: 1187-1196
Publish at: 2026-08-01

Fast-decoupled power flow optimization in 20 kV systems

10.12928/telkomnika.v24i4.27567
Abrar; Lancang Kuning University Tanjung , David; Lancang Kuning University Setiawan
Power flow optimization in electrical power distribution systems is crucial for maintaining voltage stability and energy efficiency. 20 kV distribution systems often face voltage profile instability issues due to high power losses, where conventional power flow analysis methods are sometimes inefficient in handling load fluctuations in medium-voltage networks. This study proposes power flow calculation optimization using the fast-decoupled approach to evaluate and improve voltage profiles quickly and accurately, as well as integrating reactive power compensation through strategic capacitor bank placement. Simulation results show that under existing conditions, the system experiences significant voltage drops, especially at Bus 27 (17.740 kV) and Bus 53 (7.640 kV). After implementing a 2×900 kVAr capacitor bank, the voltage profile increased dramatically: Bus 27 rose to 19.160 kV and Bus 52 reached 19.020 kV. This proves that the integration of the fast decoupled method and reactive power compensation is effective in minimizing voltage deviation and improving the operational stability of the distribution system.
Volume: 24
Issue: 4
Page: 1440-1448
Publish at: 2026-08-01

Behavioral fingerprints: driver profiling using transformer models on next generation simulation trajectory data

10.12928/telkomnika.v24i4.27790
Mohamed; Abdelmalek Essaadi University Laamimach , Mghari; Abdelmalek Essaadi University Mohammed , Aziz; Abdelmalek Essaadi University Mabrouk
Characterizing individual driver behavior is essential for advancing intelligent transportation systems (ITS) and autonomous vehicle safety. While deep learn ing models excel at macroscopic traffic prediction, individual driving styles are often aggregated away. This paper addresses this gap by proposing a novel, weakly supervised transformer framework for driver behavior profiling using high-resolution next generation simulation (NGSIM) US-101 trajectory data. We extract microscopic behavioral features including acceleration, lane change dynamics, and headway management from 30-second observation segments. A transformer encoder learns complex temporal dependencies to classify drivers into ’aggressive’ and ’normal’ profiles, achieving a 97% F1-score on proxy labeled segments. Crucially, these “proxy labels” are derived from heuristic statistics, meaning the model is trained to learn the mapping from sequences to these behavioral indicators rather than identifying objective aggression. Our methodology enables the creation of precise “behavioral fingerprints” that cap ture individual driving nuances. These insights are vital for developing adaptive ITS that anticipate traffic stability issues and enhance autonomous vehicle safety by predicting human intent.
Volume: 24
Issue: 4
Page: 1168-1176
Publish at: 2026-08-01

Experimental validation of a low-cost microcontroller-based rack-level thermal control prototype

10.12928/telkomnika.v24i4.27838
Wandercleiton; University of Genoa Cardoso , Danyelle; University of Pavia Santos Ribeiro , Thiago Augusto; Federal Institute of Espírito Santo Pires Machado , Saulo Alexandre; Bastianello Consultancy Inacio , Elielton; Hidrovias do Brasil A. Cometti , Marcelo; Federal Institute of Espírito Santo Margon , Fernando; uConnect Telecom Baptista dos Santos Neves
The rapid growth of data centers (DCs), driven by digital transformation and the increasing adoption of artificial intelligence (AI), has intensified challenges related to thermal management and operational reliability. Cooling systems account for a substantial portion of total energy consumption and often show limited effectiveness in mitigating localized hotspots and dynamic temperature variations in high-density server environments. This study presents the development and experimental validation of a low-cost, microcontroller-based (MCU-based) localized thermal control system. The proposed architecture integrates a temperature sensor, an Arduino-based control unit, pulse-width modulation (PWM) driven fan actuation, and Ethernet communication for remote monitoring. The system was implemented in a standard 19-inch rack under controlled laboratory conditions using a simulated thermal load. Experimental results, based on the average of five independent tests, demonstrated that combined operation of the prototype with rack ventilation reduced the cooling time from 45 °C to 40 °C to 50 ± 2 s, compared to approximately 5 minutes with rack ventilation alone and more than 12 minutes under natural convection. The corresponding cooling rates were 0.10 °C/s, 0.015 °C/s, and 0.007 °C/s. These results indicate that simple, distributed thermal control strategies can effectively mitigate localized overheating and support rack-level thermal stability in data center microenvironments.
Volume: 24
Issue: 4
Page: 1131-1142
Publish at: 2026-08-01

Comparative evaluation of classical and machine learning methods for medical image enhancement

10.12928/telkomnika.v24i4.27700
Md.; University of Frontier Technology Mehedi Hasan , Sujon; University of Frontier Technology Chandra Sutradhar , Zannatul; University of Frontier Technology Ferdushie , Rabeya; University of Frontier Technology Basri
Medical imaging is critical for diagnostic accuracy, yet raw images often suffer from noise and low contrast. This study provides a comparative evaluation of classical methods, namely the Laplace transform (LT), Sobel operator (SO), and histogram equalization (HE), against a data-driven convolutional neural network (CNN) using the musculoskeletal radiographs (MURA) and Human Metapneumovirus (HMPV) lung computed tomography (CT) datasets. While quantitative analysis shows that HE and SO significantly outperform other methods in isolated contrast enhancement and edge definition, they often introduce artifacts. In contrast, the CNN based approach demonstrates superior detail preservation and entropy, offering a more balanced and adaptive solution for diverse diagnostic requirements. Our findings statistically validate that although classical operators remain highly effective for specific boundary detection tasks, machine learning (ML) frameworks provide the most robust performance for cross-modality image enhancement, bridging the gap between raw data acquisition and clinical interpretation.
Volume: 24
Issue: 4
Page: 1331-1341
Publish at: 2026-08-01

A vector-valued PDE model unifying anisotropic diffusion and shock filter for color images denoising and deblurring

10.12928/telkomnika.v24i4.27726
Said; University of Sciences and Technology of Oran Mohamed Boudiaf Karoui , Salim; University of Sciences and Technology of Oran Mohamed Boudiaf Bettahar
This paper proposes a novel and efficient method for the restoration of color images degraded by the combined effects of noise and blur. To address this challenging inverse problem, a mixed partial differential equation (PDE) model based on a semi-implicit numerical scheme is developed for color image restoration. The proposed approach integrates anisotropic diffusion and shock filtering within a unified vector-valued framework, allowing the different color channels to be processed jointly while preserving their mutual correlation. This formulation enables selective smoothing that efficiently suppresses noise in homogeneous regions while simultaneously enhancing and preserving significant edge and structural features. Extensive experimental results on various color images demonstrate that the proposed method consistently outperforms existing color image restoration techniques. Quantitative evaluations based on standard image quality assessment metrics, including peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM), confirm the superior restoration performance of the proposed approach. Furthermore, the proposed model maintains chromatic consistency across color channels during simultaneous denoising and deblurring, effectively preventing the introduction of color artifacts and ensuring visually faithful restoration results. The numerical implementation of the proposed method is based on a semi-implicit scheme.
Volume: 24
Issue: 4
Page: 1342-1352
Publish at: 2026-08-01

XGBoost modeling for sparse spare-parts demand forecasting

10.12928/telkomnika.v24i4.27777
Brian Qaedi; Institut Teknologi Sepuluh Nopember (ITS) Laksono Putra , Jerry Dwi; Institut Teknologi Sepuluh Nopember (ITS) Trijoyo Purnomo
Spare parts demand in many industrial systems is inherently sparse and intermittent. In practice, long periods of zero usage are common, even though inventory must still be maintained to ensure operational reliability. This situation increases holding costs and the risk of obsolescence, while also limiting the effectiveness of conventional forecasting techniques. This study demonstrates that a global XGBoost model trained across multiple spare-part items significantly outperforms item-specific models under sparse demand conditions. Using six years of historical spare-parts usage and procurement data from the energy sector, this study compares global and single-item extreme gradient boosting (XGBoost) modeling strategies. Forecast accuracy is evaluated using mean absolute error (MAE), root mean squared error (RMSE), mean absolute percentage error (MAPE), and median absolute error (MdAE), which is particularly suitable for zero-inflated demand patterns. The results consistently show that the global XGBoost model achieves lower errors across all metrics. In particular, the global model attains a markedly lower MdAE (0.00018), indicating greater robustness when demand is irregular and intermittent.
Volume: 24
Issue: 4
Page: 1287-1293
Publish at: 2026-08-01

Adaptive trading system for sustainable forex markets

10.12928/telkomnika.v24i4.27479
Joni; Universitas Trisakti Fat , Parwadi; Universitas Trisakti Moengin , Pudji; Universitas Trisakti Astuti , Sally; Universitas Trisakti Cahyati
This study presents a sustainable and ethically aligned algorithmic trading system for the Euro/United States Dollar (EURUSD) currency pair, integrating reinforcement learning (RL) with a Sugeno-type fuzzy inference mechanism. The framework emphasizes responsible AI principles by combining adaptability and interpretability to support transparent and explainable financial decision-making. Historical EURUSD M15 data from 2020 to 2023 were used for training, while 2024 data served for out-of sample testing. The system employs EMA50-based state classification, tabular state–action–reward–state–action or SARSA learning, and a fuzzy logic layer comprising 27 expert-defined rules. During backtesting, the agent executed 785 trades, achieving a net profit of USD 61.85, a profit factor of 1.04, and a balanced win–loss ratio. Risk-adjusted analysis showed moderate resilience (sharpe ratio = 0.53) and a maximum drawdown of 59.74%. The model demonstrated strong equity stability (ESI = 0.9672) and sensitivity to macroeconomic events identified through cumulative sum (CUSUM) analysis. While the system maintained capital preservation and interpretability, responsiveness under volatile conditions requires improvement. Future work will focus on adaptive exit logic, volatility-aware reward mechanisms, and regime-sensitive policy optimization. This study contributes to advancing sustainable, transparent, and risk-aware artificial intelligence (AI) frameworks in algorithmic trading.
Volume: 24
Issue: 4
Page: 1267-1277
Publish at: 2026-08-01

Beyond reductionism: systems thinking for the next generation of electrical and computer engineering

10.12928/telkomnika.v24i4.3776
Tole; Universitas Ahmad Dahlan Sutikno
Classical electrical and computer engineering has achieved remarkable progress through reductionist methodologies that decompose complex systems into manageable, analyzable, and optimizable components. While this paradigm remains indispensable for scientific rigor and engineering design, it is increasingly challenged by contemporary systems characterized by interconnectedness, dynamic interactions, and multi-scale complexity. This editorial argues that future engineering requires extending, rather than replacing, reductionist thinking with systems thinking capable of capturing interdependence, emergence, resilience, and holistic system behaviour. Beyond component-level optimization, engineering must increasingly consider interactions among technological, human, environmental, and societal dimensions that collectively shape system performance and long term sustainability. Systems thinking therefore provides a complementary paradigm for understanding how complex engineering systems adapt, evolve, and generate behaviours that cannot be inferred solely from individual subsystems. This perspective redefines electrical and computer engineering as an integrated socio-technical discipline in which analytical precision is combined with systemic understanding to address increasingly complex real-world challenges. Moving beyond reductionism does not diminish the value of analytical methods but expands their scope within broader interconnected contexts. This paradigm shift establishes the conceptual foundation for the subsequent evolution toward adaptive, human centred, and ultimately responsible engineering.
Volume: 24
Issue: 4
Page: 1083-1090
Publish at: 2026-08-01

Reconfigurability of graphene-based hexagonal patch operating in the Ku band

10.12928/telkomnika.v24i4.27737
Hassna; Moulay Ismail University Agoumi , Seddik; Moulay Ismail University Bri , Youssef; Mohammed V University El Amraoui , Adil; Moulay Ismail University Saadi
This paper presents a novel graphene-based impedance reconfigurability approach for Ku-band hexagonal patch antennas, demonstrated at both the single-element and 4×4 array levels. Unlike conventional metallic or diode based reconfigurable antennas, frequency tuning is achieved by exploiting the tunable surface conductivity of graphene integrated into E-shaped slots, without altering the antenna geometry or employing active lumped components. Antenna performance is evaluated using full-wave electromagnetic simulations for two graphene states (“on” state and “off” state), representing distinct surface impedance conditions. The single element exhibits dual resonances at 14.81 GHz and 15 GHz, with reflection coefficients of -52.56 dB and -38.18 dB, bandwidths of 783 MHz and 517 MHz, and a peak gain of 7.8 dBi. The 4×4 array exhibits multiple resonances between 12.5–16.8 GHz (off state) and 12.6–17.15 GHz (on state), achieving bandwidths up to 2750 MHz and a maximum gain of 13.64 dBi. These results demonstrate a scalable, geometry-preserving reconfigurable antenna solution for compact Ku-band systems.
Volume: 24
Issue: 4
Page: 1372-1384
Publish at: 2026-08-01

Mapping research trends on tropical cyclone–induced flood susceptibility: a bibliometric and systematic review method

10.12928/telkomnika.v24i4.27614
Soenardi; IPB University Soenardi , Bambang; IPB University Dwi Dasanto , Yonny; IPB University Koesmaryono , I; IPB University Putu Santikayasa , Giarno; Agency of Meteorology, Climatology, and Geophysics Giarno
Climate change has intensified tropical cyclones (TC), increasing extreme rainfall and flood hazards in many regions. Flood susceptibility (FS) mapping is therefore essential for understanding flood risk. This study analyzes global research trends on TC-induced FS by integrating bibliometric analysis and a preferred reporting items for systematic reviews and meta-analyses (PRISMA)-based systematic literature review (SLR) using Google Scholar (GS) publications from 2014 to 2024. A total of 993 journal articles were analyzed, yielding an h-index of 101 and a g-index of 168, indicating strong and growing research interest. The results reveal an increasing application of machine learning (ML), deep learning (DL), remote sensing (RS), and geographic information systems (GIS) for FS mapping. Several gaps remain, including limited use of high-resolution data, underrepresentation of data-scarce and equatorial regions, restricted integration of hybrid models, and a lack of long-term assessments considering climate change and socio-economic factors. The model’s performance is also highly dependent on data quality and regional characteristics, limiting its generalizability across different conditions. The main contribution of this study is the knowledge mapping and synthesis of TC-induced FS research, providing a structured foundation for future studies and supporting evidence-based flood risk management and climate adaptation.
Volume: 24
Issue: 4
Page: 1253-1266
Publish at: 2026-08-01
1 2 3
...
2059 2060
Show 7 of 2060

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