2026-05-08T14:30:12Z https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/oai

oai:sycom.lab-lamis-univ-tebessa.dz:article/3 2025-07-21T12:47:55Z sycom:ART driver

Unveiling Identity Through Anatomy: Person Verification Using Vision Transformers on Chest X-Rays Radiographs Farah, Hazem Bennour, Akram Afrin, Syeda Sadia Soltani, Hama Adjal, Ali Chest X-rays, Cosine similarity, Person verification, Triplet Neural Network, Vision transformers (Vit). Context: The prospective utilization of medical imaging data for reliable individual identification and authentication has garnered significant interest in both security and healthcare sectors. This importance is particularly amplified during disaster scenarios, where conventional means of human verification become ineffective. In these challenging conditions, Chest X-rays serve as an essential resource by capturing unique anatomical details of the rib cage, lungs, and heart features that persist as reliable verification even when the body is compromised. Objective: We propose the creation of an innovative verification system for image retrieval, specifically designed to enhance person verification using chest X-ray images. Method: The system integrates a deep learning paradigm, leveraging Triplet network architecture; while uniquely use cosine similarity as a metric to assess similarity and dissimilarity between image features. Such an approach enables a more nuanced and robust feature comparison, leading to improved retrieval accuracy and verification performance. Building upon this premise, we propose the creation of an innovative verification system for image retrieval. Notably, our framework employs a state-of-the-art Vision Transformer (ViT) as the backbone for the Triplet network. Results: The ViT backbone offers robust capabilities in extracting and contextualizing features, thereby enhancing the discriminative power of the triplet architecture and ensuring improved retrieval accuracy. This novel integration not only expands the existing toolkit for medical image analysis but also reinforces the reliability of identity verification systems. Conclusions: The dual use of geometric and angular similarity measures, coupled with the advanced feature extraction of the ViT, offers a precise and dependable solution, particularly in high-stakes scenarios such as emergencies and security-critical applications. Laboratory of mathematics, informatics and systems (LAMIS), Echahid Cheikh Larbi Tebessi University- Tebessa, Algeria 2025-07-19 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/3 10.64409/sycom.v1.i1.3 Systems and Computing; Vol. 1 No. 1 (2025) 3088-7941 eng https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/3/5 Copyright (c) 2025 Hazem Farah, Akram Bennour, Syeda Sadia Afrin, Hama Soltani, Ali Adjal (Author) https://creativecommons.org/licenses/by/4.0

oai:sycom.lab-lamis-univ-tebessa.dz:article/4 2025-07-21T12:45:37Z sycom:ART driver

Empowering Biometric Security: A Paradigm Shift in Remote Identification Belhocine, Yacine Meraoumia, Abdallah Bendjenna, Hakim Saiga, Mohammed Biometric authentication, Dictionary learning, FKP, Fuzzy vault, Identity protection, Secure remote identification. Context: In an increasingly digital world, the need for secure and reliable identity verification is more critical than ever. Biometric authentication stands out due to its inherent uniqueness and resistance to forgery, yet its performance is still affected by real-world challenges such as image noise and lighting inconsistencies. Objective: This paper aims to enhance the robustness and accuracy of biometric systems, specifically for remote authentication scenarios, by addressing the limitations posed by environmental variations. Method: The proposed solution is a novel FKP-based biometric authentication system that employs a Triple Texture Feature Extraction (TTFE) technique to capture detailed information from both spatial and frequency domains. To further improve recognition performance, Projective Dictionary Learning is used to refine the feature representation. For secure data handling, a fuzzy vault scheme is integrated to encrypt biometric templates using secret keys, allowing secure authentication over potentially untrusted networks. Results: Experiments conducted on standard benchmark datasets demonstrate significant improvements in both accuracy and resilience to challenging conditions such as noise and lighting variations. Conclusions: This integrated approach successfully enhances the reliability and security of biometric authentication systems, paving the way for more practical and scalable real-world deployments. Laboratory of mathematics, informatics and systems (LAMIS), Echahid Cheikh Larbi Tebessi University- Tebessa, Algeria 2025-07-19 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/4 10.64409/sycom.v1.i1.4 Systems and Computing; Vol. 1 No. 1 (2025) 3088-7941 eng https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/4/4 Copyright (c) 2025 Yacine BELHOCINE, Abdallah Meraoumia, Hakim Bendjenna, Mohammed Saigaa (Author) https://creativecommons.org/licenses/by/4.0

oai:sycom.lab-lamis-univ-tebessa.dz:article/5 2025-07-21T12:41:02Z sycom:ART driver

Attention-Gateway U-Net for Mammographic Mass Segmentation: An Empirical Study Soltani, Hama Islam, Yousuf Farah, Hazem Bendib, Issam Haouam, Mohamed-Yassine Amroune, Mohamed Attention Gateway U-net, Breast Cancer, Mammography, Segmentation. Context: Accurate segmentation of breast masses in mammographic images is a critical step for early breast cancer detection and effective clinical decision-making. However, this task remains challenging due to variability in mass appearance, low contrast with surrounding tissues, and complex anatomical structures. Objective: This study proposes a novel deep learning framework based on an attention-guided U-Net architecture, specifically designed for the segmentation of suspicious masses in mammograms. The primary objective is to investigate the impact of integrating attention gates into the U-Net structure and to evaluate how these mechanisms influence the accuracy and robustness of mass segmentation. Method: Attention gates are embedded within the skip connections and function by dynamically highlighting relevant features from the encoder while suppressing less informative regions. This selective focus enables the model to more accurately delineate the often subtle boundaries of breast masses. Results: The proposed model is evaluated on the publicly available INbreast mammography dataset, and its performance is compared against the baseline U-Net and other state-of-the-art segmentation methods. Quantitative results demonstrate that the attention-enhanced U-Net significantly outperforms its counterparts in segmentation accuracy, particularly in challenging cases involving dense breast tissue or ill-defined masses. Conclusions: This study highlights the effectiveness of attention mechanisms in enhancing mammographic mass segmentation and represents a valuable step toward more intelligent and reliable Computer-Aided Diagnosis (CAD) systems for breast cancer. Laboratory of mathematics, informatics and systems (LAMIS), Echahid Cheikh Larbi Tebessi University- Tebessa, Algeria 2025-07-19 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/5 10.64409/sycom.v1.i1.5 Systems and Computing; Vol. 1 No. 1 (2025) 3088-7941 eng https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/5/1 Copyright (c) 2025 Hama Soltani, Yousuf Islam, Hazem Farah, Issam Bendib, Mohamed-Yassine Haouam, Mohamed Amroune (Author) https://creativecommons.org/licenses/by/4.0

oai:sycom.lab-lamis-univ-tebessa.dz:article/6 2025-07-19T22:51:58Z sycom:ART driver

Towards Accurate Detection and Classification of Skin Cancer Using AI-Powered Image Analysis Benaliouche, Houda Guelib, Ikram Slimani, Yahia Salhi, Dalel Guerfi, Younes Convolutional Neural Networks (CNNs), Deep learning, Image classification, Medical imaging, Skin cancer detection. Context: Skin cancer is one of the most common and lethal forms of cancer if not diagnosed early. Detection at an early stage can mean a world of difference in survival. Despite advances in technology, most areas continue to face a lack of available dermatology specialists, making timely diagnosis a challenge. This gap between the need for early detection and the limited availability of experts highlights the urgency for automated, reliable diagnostic tools. Objective: In response to this need, the purpose of this work is to create a fast and accurate system for skin cancer detection that distinguishes between malignant and benign lesions. Method: This study, based on a Convolutional Neural Network (CNN), addresses some challenges, including limited data, image quality issues, and classification. Results: A balanced training set was used to train the model, which achieved a test accuracy of 88.60% and a macro-averaged Receiver Operating Characteristic Area Under the Curve (ROC-AUC) score of 0.95. Other significant metrics, such as precision, recall, and F1-score also validated the performance of the model. We developed a web-based interface to enable practical deployment and usability for real-world applications. Conclusions: The findings demonstrate the potential of deep learning techniques to enhance skin cancer detection and serve as a point of reference for future research on multi-class classification and real-time diagnostic platforms. Laboratory of mathematics, informatics and systems (LAMIS), Echahid Cheikh Larbi Tebessi University- Tebessa, Algeria 2025-07-19 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/6 10.64409/sycom.v1.i1.6 Systems and Computing; Vol. 1 No. 1 (2025) 3088-7941 eng https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/6/2 Copyright (c) 2025 Houda Benaliouche, Ikram Guelib, Yahia Slimani, Dalel Salhi, Younes Guerfi (Author) https://creativecommons.org/licenses/by/4.0

oai:sycom.lab-lamis-univ-tebessa.dz:article/14 2025-07-21T12:44:23Z sycom:ART driver

Customer Churn Prediction in Neobanking System Using Predictive Analytics and Feature Selection Babatunde, Abdulrauph Olanrewaju Yinusa, Sheriffdeen Ade Oladipo, Idowu Dauda Asaju-Gbolagade, Ayisat Wuraola Classification, Customer Churn, Churn prediction, Data Mining Context: Customer behavior, including loyalty and satisfaction, is increasingly volatile due to rapid technological and market changes. Neobank startups, in particular, face significant challenges related to customer churn, which can severely impact profitability and reputation. Objective: To develop a predictive model that identifies customers at risk of churning in the neobanking sector, enabling proactive retention strategies. Method: This study employed data mining techniques using three classification algorithms, Logistic Regression, Naïve Bayes, and Decision Tree, implemented in the WEKA platform. Feature selection methods based on accuracy, precision, and correlation were applied to identify key churn indicators. Results: The Decision Tree algorithm outperformed the others, achieving an accuracy of 80.5%. It demonstrated superior performance, particularly when all features were included in the model. Key predictive features were successfully identified through feature selection techniques. Conclusions: The findings confirm that Decision Trees are effective in predicting customer churn in neobanks. Understanding and targeting the key factors influencing churn can help neobanks retain customers and maintain a competitive edge. Laboratory of mathematics, informatics and systems (LAMIS), Echahid Cheikh Larbi Tebessi University- Tebessa, Algeria 2025-07-19 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/14 10.64409/sycom.v1.i1.14 Systems and Computing; Vol. 1 No. 1 (2025) 3088-7941 eng https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/14/3 Copyright (c) 2025 Abdulrauph Olanrewaju BABATUNDE, Sheriffdeen Ade YINUSA, Idowu Dauda OLADIPO, Ayisat Wuraola ASAJU-GBOLAGADE (Author) https://creativecommons.org/licenses/by/4.0

oai:sycom.lab-lamis-univ-tebessa.dz:article/15 2026-01-15T06:38:52Z sycom:ART driver

Systems Thinking and Systems Dynamics Approach to Demand Side Management Nagoma, Donnell Joe Kameswara Musti Demand Side Management Vensim System Dynamics Systems Thinking Price Elasticity Electricity Demand Consumer behaviour Context: Demand Side Management (DSM) provides utilities with tools to modify end users’ consumption patterns, helping reduce peak demand, alleviate stress on the power grid, and even support with the integration of renewables. Nonetheless, DSM is an inherently complex and interdependent system stretching across numerous factors such as economic, behavioural, and technological with different feedback dynamics and competing incentives. To address this complexity a Systems Thinking and Systems Dynamics approach is applied. Objective: The aim of this paper is to develop a quantitative Systems Dynamics model in the Vensim simulation environment, capturing the dynamic transition between short run and long run price elasticities. Methods: The model defines variables such as baseline demand, electricity price, and both short-run and long-run elasticities. A stock variable, Effective Price Elasticity, is used to dynamically represent the shift from short-run to long-run responses. Simulation parameters were gathered from literature. Results: Simulations demonstrate that the model confirms expected behaviour, showing an initial less elastic response to the price changes followed by a more notable gradual adjustment over time as the effective elasticity converges to its long-run value. Conclusion: This model can provide a framework for simulating the dynamic interplay between electricity price and demand. Its ability to simulate short-run and long-run elasticity effects offers valuable insights for energy planners and policy makers. Laboratory of mathematics, informatics and systems (LAMIS), Echahid Cheikh Larbi Tebessi University- Tebessa, Algeria 2026-01-15 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/15 10.64409/sycom.v2.i1.15 Systems and Computing; Vol. 2 No. 1 (2026) 3088-7941 eng https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/15/8 https://creativecommons.org/licenses/by/4.0

oai:sycom.lab-lamis-univ-tebessa.dz:article/22 2026-01-15T06:38:52Z sycom:ART driver

Avoiding Local Minima in Overlap-Based Clustering by Random Swap Wang, Fei Li, Le Fränti, Pasi Clustering Overlap k-means Cluster overlap Random swap Context: Cluster overlap has been recently introduced as an alternative objective function to the traditional sum-of-squared error (SSE) function used in k-means. It estimates the overlap by counting how many points are shared between the neighbor clusters. The overlap k-means was shown to provide more stable centroid locations than SSE, but at the cost of losing the dynamics of the k-means algorithm. Objective: In this paper, we address this drawback by adding a random swap step to the algorithm and introducing a new Overlap random swap. Method: It consists of three steps: (1) selective replacement of the centroids, (2) standard cluster assignment step, (3) weighted centroid calculation giving less weight to the overlapping points. Results: Experimental results confirm that ORS achieves competitive clustering accuracy, stable centroid locations, and improved boundary separation. The clustering accuracy (ACC) is about 95% and the centroid index (CI) is about 0.1. The algorithm converges within 3000 swap iterations. Conclusions: These results indicate that ORS helps avoid local minima and maintains the robustness of overlap-based clustering. Laboratory of mathematics, informatics and systems (LAMIS), Echahid Cheikh Larbi Tebessi University- Tebessa, Algeria 2026-01-15 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/22 10.64409/sycom.v2.i1.22 Systems and Computing; Vol. 2 No. 1 (2026) 3088-7941 eng https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/22/6 https://creativecommons.org/licenses/by/4.0

oai:sycom.lab-lamis-univ-tebessa.dz:article/27 2026-01-15T06:38:52Z sycom:ART driver

Developing an Integrated HMSAM-FoMO Model and Its Instrument Testing for Social Live Streaming Yudhanta, Satya Durachman, Yusuf Fitroh, Fitroh Zulfiandri Subiyakto, Aang Model Development HMSAM Fear of Missing Out Social Live Streaming Instrument Validation Context: The rapid growth of social live streaming platforms has transformed how individuals interact and experience digital entertainment. Despite this expansion, prior studies have largely focused on commercial or gaming-related live streaming, leaving the social and hedonic aspects underexplored. Furthermore, psychological mechanisms such as the Fear of Missing Out (FoMO) have received limited attention in explaining behavioral intention in live streaming environments. Objective: This study aims to develop and validate an integrated model combining the Hedonic-Motivation System Adoption Model (HMSAM) with FoMO to explain users’ behavioral intention in social live streaming contexts. Method: A quantitative approach was employed using Partial Least Squares Structural Equation Modeling (PLS-SEM) with data from 57 Indonesian social media users. The research process consisted of three stages: conceptual integration, instrument operationalization, and empirical validation, supported by expert qualitative evaluation to enhance construct clarity and theoretical consistency. Results: The findings show that hedonic motivations such as joy, curiosity, and immersion strongly influence behavioral intention, while FoMO acts both as a direct determinant and as a mediator linking hedonic motivations to user engagement. Two indicators (CTR2 and CTR4) were removed for weak conceptual fit, and constructs with AVE values slightly below 0.50 (Control and Perceived Usefulness) were retained for theoretical reasons. The integrated HMSAM–FoMO model explained 64.4% of the variance in behavioral intention, demonstrating strong explanatory power. Conclusions: The study contributes theoretically by extending HMSAM with FoMO to capture both emotional enjoyment and psychological tension in digital engagement. It also provides practical guidance for designing psychologically balanced live streaming experiences that enhance enjoyment while minimizing FoMO-induced anxiety. These findings offer a foundation for future research and policy development toward healthier digital participation. Laboratory of mathematics, informatics and systems (LAMIS), Echahid Cheikh Larbi Tebessi University- Tebessa, Algeria 2026-01-15 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/27 10.64409/sycom.v2.i1.27 Systems and Computing; Vol. 2 No. 1 (2026) 3088-7941 eng https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/27/10 https://creativecommons.org/licenses/by/4.0

oai:sycom.lab-lamis-univ-tebessa.dz:article/28 2026-01-15T06:38:52Z sycom:ART driver

Recognition of GUI Web Elements using the Deep Learning Approach and Yolo Architecture Prifti, Helga Bekir Karlik Deep Learning GUI Web Element YOLOv8 Agile Framework SDLC Context: The growing complexity of modern web and mobile applications, shaped by rapid UI/UX advancements and agile software development practices, demands efficient and automated recognition of graphical user interface (GUI) elements, as manual testing remains costly and time-consuming. Objective: This study aims to evaluate the performance of the YOLOv8 deep learning architecture for accurate recognition of diverse GUI components, such as buttons, fields, headings, links, and images, from real-world application screenshots. Method: A small YOLOv8 model was trained on the Roboflow GUI element detection dataset containing over 1,000 annotated website screenshots, using 35 training epochs, a batch size of 8, and an image resolution of 640×640. Results: The model achieved measurable improvements, with precision rising from 0.368 to 0.454, recall increasing from 0.296 to 0.425, and mAP50–95 improving from 0.101 to 0.232. Strongest detection was observed for buttons and input fields, while weaker performance was noted for iframes and labels due to their inherent ambiguity. Conclusions: The results demonstrate that YOLOv8 has significant potential in automating GUI recognition, reducing reliance on manual testing in agile workflows, and improving interface validation. Further optimization with larger datasets and advanced augmentation methods is recommended to enhance robustness and generalization Laboratory of mathematics, informatics and systems (LAMIS), Echahid Cheikh Larbi Tebessi University- Tebessa, Algeria 2026-01-15 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/28 10.64409/sycom.v2.i1.28 Systems and Computing; Vol. 2 No. 1 (2026) 3088-7941 eng https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/28/11 https://creativecommons.org/licenses/by/4.0

oai:sycom.lab-lamis-univ-tebessa.dz:article/29 2026-01-15T06:38:52Z sycom:ART driver

Improving Arabic Dialect Processing in IoT Systems: A Comparative Study of Baseline and Dialect-Aware AI Models Djehaiche, Rania Djehaiche, Yamena Arabic dialects IoT Dialect-Aware model Baseline model AI Context: Bringing voice-controlled interfaces into Internet of Things (IoT) systems has created fresh opportunities for smart environments. However, existing voice assistants often struggle with non-standardized languages, especially Arabic dialects. Objective: This research paper explores the challenges and potential of integrating five Arabic dialect variants, namely Modern Standard Arabic (MSA), known as Fusha (الفصحى), Egyptian, Levantine, Gulf, and Algerian dialects, into AI-driven IoT systems. Methods: For each dialect, a comparative simulation was performed using two AI models: a baseline model and a dialect-aware model. Key simulated metrics included automatic speech recognition (ASR) accuracy, intention recognition, task success rate, and system response time. Results: The results consistently show that the dialect-aware model outperforms the baseline model in all metrics. It provides higher ASR and intention recognition accuracy, improved task success rates, and faster response times, especially for regional dialects. The Algerian dialect, while still challenging, benefited significantly from the dialect-aware adaptations of the improved model. These results highlight the potential of dialect-aware AI to close the performance gap caused by linguistic variation and code-switching. Conclusion: This study highlights the importance of considering linguistic diversity when developing accessible, culturally appropriate IoT interfaces that ensure a more inclusive and natural user interaction. Laboratory of mathematics, informatics and systems (LAMIS), Echahid Cheikh Larbi Tebessi University- Tebessa, Algeria 2026-01-15 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/29 10.64409/sycom.v2.i1.29 Systems and Computing; Vol. 2 No. 1 (2026) 3088-7941 eng https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/29/9 https://creativecommons.org/licenses/by/4.0

oai:sycom.lab-lamis-univ-tebessa.dz:article/30 2026-01-15T06:38:52Z sycom:ART driver

Impact of Divergence Angle on FSO System Performance: A Study of SNR, BER, and Q Factor Allaoua, Oumaima Djellab, Hanane Maamri, Fouzia Belhocine, Yacine Saidi, Riad Boumehrez, Farouk Sahour, Abdelhakim Salem, Ahmed mohamed FSO SNR BER Divergence angle Q factor Context: Free Space Optics (FSO) transmission is a promising alternative to conventional wired networks, offering high bandwidth, low latency, and enhanced security. However, FSO performance is highly sensitive to physical factors, particularly the divergence angle of the optical beam. Objective: This study aims to evaluate the impact of beam divergence on key FSO system performance metrics to guide optimal system design. Method: MATLAB-based simulations were conducted to model Signal-to-Noise Ratio (SNR), Bit Error Rate (BER), and Q factor under varying divergence angles. The sensitivity of system performance to beam alignment and signal strength was analyzed across different divergence configurations. Results: Results indicate that smaller divergence angles provide higher SNR and Q factor values and lower BER, ensuring superior communication quality but requiring precise alignment. Conversely, larger divergence angles tolerate misalignment more effectively but at the cost of reduced signal quality. Conclusion: These findings emphasize the critical role of optimizing beam divergence to balance alignment tolerance and signal integrity, thereby enhancing the efficiency and reliability of optical wireless communication systems. Laboratory of mathematics, informatics and systems (LAMIS), Echahid Cheikh Larbi Tebessi University- Tebessa, Algeria 2026-01-15 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/30 10.64409/sycom.v2.i1.30 Systems and Computing; Vol. 2 No. 1 (2026) 3088-7941 eng https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/30/13 https://creativecommons.org/licenses/by/4.0

oai:sycom.lab-lamis-univ-tebessa.dz:article/33 2026-01-15T06:38:52Z sycom:ART driver

Benchmarking New Adaptive Equalizers Leveraging a Different Robust Adaptive Algorithms: NLMS, APA, PAP, and MRNQ Lounoughi, Fatima DJENDI , Mohamed Digital communication system Decision feed forward equalizer Convergence speed Computational complexity NLMS algorithm APA PAP algorithm MRNQ algorithm Context: In the literature, numerous adaptive equalization techniques have been developed to effictively reduce inter-symbol interference in digital communication systems. Objective: This study conducts a comparative analysis of four such adaptive algorithms: NLMS, APA, PAP, and MRNQ, implemented within a decision feed-forward equalizer (DFE) stucture. The purpose of this evaluation is to identify the strengths and limitations of each algorithm under the same transmission conditions. The insights gained from this comparaison can guide the choice of equalization strategy for practical implementations in modern communication systems. Method: The performance of each equalizer is assessed using multiple evaluation criterion, including computational complexity, constellation diagram analysis, the Nyquist criterion, and mean squared error (MSE) performance. Results: All four algorithms effectively reduce inter-symbol interference and ensure proper signal equalization. However, DFE-PAP and APA-DFE show superior speed and accuracy, MRNQ-DFE offers a good trade-off, and NLMS-DFE performs the weakest. Conclusions: In summary, the study highlights DFE-PAP as the most efficient and pratical solution, combining strong ISI mitigation with low computationnal cost. While all algorithms perform well, DFE-PAP and APA-DFE demonstrate clear advantages in convergence speed and accuracy.  Laboratory of mathematics, informatics and systems (LAMIS), Echahid Cheikh Larbi Tebessi University- Tebessa, Algeria 2026-01-15 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/33 10.64409/sycom.v2.i1.33 Systems and Computing; Vol. 2 No. 1 (2026) 3088-7941 eng https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/33/12 https://creativecommons.org/licenses/by/4.0

oai:sycom.lab-lamis-univ-tebessa.dz:article/35 2026-01-15T06:38:52Z sycom:ART driver

Evaluation of SE, ECA, and CA Modules in ResNet-50 for the Classification of Grape Leaf Diseases Mekhilef, khedidja Hemam, Mounir Classfication Attention mechanisms Agriculture Plant diseases Context: Plant diseases pose a major threat to agricultural production, as they hinder crop growth and significantly reduce yields. In this context, the automatic detection of diseases using artificial intelligence techniques represents an effective solution. Objective: This work presents a comparative study on the integration of attention mechanisms into the ResNet-50 architecture, applied to the classification of grape leaf diseases. Method: Three attention modules were considered: SE (Squeeze-and-Excitation), ECA (Efficient Channel Attention), and CA (Coordinate Attention), each integrated separately into the model to evaluate its impact. Results: The performance of the different configurations was assessed by considering the quality of the results, the model complexity, and the processing speed. Experimental results show that the ECA module achieved the best performance with a reduced number of parameters, followed by SE and then CA. Conclusion: This study highlights the contribution of attention mechanisms in improving the effectiveness of CNN architectures for plant disease detection. Laboratory of mathematics, informatics and systems (LAMIS), Echahid Cheikh Larbi Tebessi University- Tebessa, Algeria 2026-01-15 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/35 10.64409/sycom.v2.i1.35 Systems and Computing; Vol. 2 No. 1 (2026) 3088-7941 eng https://sycom.lab-lamis-univ-tebessa.dz/index.php/sycom/article/view/35/7 https://creativecommons.org/licenses/by/4.0