KGKamal El Guemmat

Kamal El Guemmat

*This is an algorithmically generated author record based on publicly available information.

Kamal El Guemmat has a total of 27 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    DeepShield-IoT: A Hybrid AI and Lotka–Volterra Model for Efficient IoT Intrusion Detection Systems

    by Mohamed Bachar, Azeddine Khiat, Kamal El Guemmat

    Internet of Things devices introduce significant security challenges caused by their heterogeneous and resource-constrained nature in many sectors, such as healthcare, industry, education, and agriculture. Intrusion detection systems (IDSs) serve a key role in identifying malicious activities in such environments; traditional approaches cannot often capture dynamic interactions and temporal correlations in network traffic. In this research, we propose a novel hybrid IDS technique utilizing Long Short-Term Memory (LSTM) networks in conjunction with Lotka–Volterra (LV) dynamic modeling. The LSTM component is employed to learn temporal patterns and estimate system states from IoT traffic, while the LV model captures the dynamic interaction between normal and malicious behavior. We introduce a mathematically based decision mechanism on an anomaly score for effective classification. The model’s results on DataSense: CIC IIoT dataset 2025 achieve an accuracy of 99.85%, outperforming other models, and have a detection time of 47 ms and a reduced-complexity algorithm. These results highlight the effectiveness of combining artificial intelligence with dynamic system modeling for intrusion detection in IoT environments, providing a promising direction for future research in intelligent cybersecurity systems.

  • Zero Trust Architecture in Healthcare: A Systematic Mapping Study

    by Marouane El Aatouani, Kamal El Guemmat, Abdelmajid Bousselham, Soulaimane Bahi, Mohamed Akram Hamdaoui

  • Open Access

    DeepShield-IoT: A Hybrid AI and Lotka–Volterra Model for Efficient IoT Intrusion Detection Systems

    by Mohamed Bachar, Azeddine Khiat, Kamal El Guemmat

    Internet of Things devices introduce significant security challenges caused by their heterogeneous and resource-constrained nature in many sectors, such as healthcare, industry, education, and agriculture. Intrusion Detection Systems (IDS)s serve a key role in identifying malicious activities in such environments; traditional approaches cannot often capture dynamic interactions and temporal correlations in network traffic. In this research, we propose a novel hybrid IDS technique utilizing Long Short-Term Memory (LSTM) networks in conjunction with Lotka-Volterra (LV) dynamic modeling. The LSTM component is employed to learn temporal patterns and estimate system states from IoT traffic, while the LV model captures the dynamic interaction between normal and malicious behavior. We introduce a mathematically based decision mechanism on an anomaly score for effective classification. The model’s results on the CICIIoT 2025 dataset achieve an accuracy of 99.85, outperforming other models, and have a detection time of 47 ms and a reduced-complexity algorithm. These results highlight the effectiveness of combining artificial intelligence with dynamic system modeling for intrusion detection in IoT environments, providing a promising direction for future research in intelligent cybersecurity systems.

  • Electricity Theft Detection: Experimental Comparison of Ensemble and Traditional Machine Learning Methods

    by Imane Rakine, Kamal El Guemmat, Sara Ouahabi, Issam Atouf, Mohamed Talea, Tarik Bouragba

  • Open Access

    Predictive insights into student online learning adaptability: elevating e-learning landscape

    by Mohamed El Ghali, Issam Atouf, Kamal El Guemmat, Mohamed Talea

    In Morocco’s rapidly transforming educational landscape, this study delves into students’ adaptability to online learning environments by integrating sophisticated artificial intelligence (AI) algorithms and hyperparameter optimization techniques. This research uses the comprehensive “online learning adaptivity” dataset to identify pivotal factors influencing student flexibility and effectiveness in e-learning platforms. We applied various AI models, with a particular emphasis on the CatBoost classifier, which exhibited exceptional predictive performance, achieving an accuracy rate near 98%. This high precision in predicting student adaptiveness offers essential insights into tailoring digital education systems. The results underscore the significant potential of machine learning technologies to enhance educational methodologies by catering to the diverse needs of students. Such capabilities are instrumental for educators and policymakers dedicated to refining e-learning strategies that effectively accommodate individual learning styles, ultimately improving the broader educational outcomes in Moroccan tertiary education. These findings advocate for a more nuanced understanding of the interplay between student behavior and technological solutions, providing a roadmap for developing more responsive and effective educational platforms.

  • Open Access

    Predictive maintenance approaches: A systematic literature review

    by Hicham Taoufyq, Kamal El Guemmat, Khalifa Mansouri, Fatiha Akef

    Purpose: Predictive maintenance (PdM) aims to optimize maintenance operations by detecting operational anomalies and potential equipment failures before they lead to costly unplanned downtime. The goal is to minimize reactive maintenance and reduce the frequency of preventive maintenance interventions. This paper evaluates PdM strategies using knowledge-based, physics-based, and data-driven models to propose an integrated approach that enhances prediction accuracy, aligning with Industry 4.0 goals.Design/methodology/approach: A Systematic Literature Review (SLR) is conducted to examine the strengths and weaknesses of knowledge-based, physics-based, and data-driven models in predictive maintenance. The study assesses existing research, compares methodologies, and identifies opportunities for integrating these models to improve PdM outcomes.Findings: The review indicates that no single approach—whether knowledge-based, physics-based, or data-driven—is sufficient to meet the comprehensive demands of predictive maintenance. Instead, an integrated approach that combines these three models provides more accurate and cost-effective maintenance solutions, supporting the automation and efficiency goals of Industry 4.0.Research limitations/implications: The study's findings are limited by the availability of real-world data and case studies. Future research should focus on testing the proposed integrated model in diverse industrial contexts to validate its effectiveness across different sectors.Practical implications: The proposed approach offers industries a more reliable framework for optimizing maintenance strategies, improving operational efficiency, and reducing costs associated with equipment failures and excessive preventive measures.Social implications: By enhancing predictive maintenance, the integrated model supports sustainability efforts by reducing waste, improving resource utilization, and contributing to the longevity of machinery and equipment.Originality/value: This research offers a novel contribution by integrating knowledge-based, physics-based, and data-driven models into a unified PdM approach. It provides valuable insights for both academia and industry, especially in the context of Industry 4.0.


  • Open Access

    Comprehensive Review of Intrusion Detection Techniques: ML and DL in Different Networks

    by Imane Rakine, Aziz Oukaira, Kamal El Guemmat, Issam Atouf, Sara Ouahabi, Mohamed Talea, Tarik Bouragba

    With the increasing number of new attacks, virtualized and distributed networks require greater attention and investment in cybersecurity. Organizations must rely on effective Intrusion Detection Systems (IDS) to detect both known and novel attacks. Therefore, Machine Learning (ML) and Deep Learning (DL) techniques have been widely used for intrusion detection. Several studies have reviewed ML and DL-based detection models, but they often overlook the specific networks targeted by these models. It is crucial to understand not only which methods are effective but also the contexts in which they are effective. This study aims to fill this gap by reviewing and classifying recent contributions based on their target networks. It focuses on three key network types: Cloud Computing (CC), Internet of Things (IoT), and Software-Defined Networks (SDN). Our study emphasizes the importance of thoroughly understanding the strengths and vulnerabilities of a given network, which is an important step towards developing effective ML- and DL-based intrusion detection approaches. We first provide an overview of related works and our research steps, followed by a presentation of ML and DL techniques, and commonly used datasets in this field. Next, a detailed presentation of the current research on IDS based on ML and DL techniques by network categories is provided. The strengths and limitations of ML and DL algorithms, which are frequently used for intrusion detection, are highlighted. Finally, the challenges are discussed and future research directions are proposed.


  • Open Access

    Enhancing convolutional neural network based model for cheating at online examinations detection

    by Sara Ouahabi, Rihab Aboudihaj, Nawal Sael, Kamal El Guemmat

    In the last few years, e-learning has revolutioning education, giving students access to diverse and adaptable on-line resources, but it has also face a major challenge: cheating on online exams. Students now use variant cheating methods include consulting unauthorized documents, communicating with others during the exam, searching for information on the internet. Combating these cheating practices has become crucial to preserving the integrity of academic assessments. In this context, artificial intelligence (AI) has emerged as an essential tool for mitigating this fraudulent behavior. Equipped with advanced machine learning capabilities, AI can examine a wide range of data to detect student suspicious behavior. This study develops an approach based on a convolutional neural network (CNN) model designed to detect cheating by analyzing candidates' head movements during online exams. By exploiting the FEI dataset, this model achieves an interesting accuracy of 97.28%. In addition, we compare this model to the well-known transfer learning models used in the literature namely, ResNet50, VGG16, DenseNet21, MobileNetV2, and EfficientNetB0 demonstrating the out performance of our approach in detecting cheating during online exams.

  • Open Access

    Developing digital capabilities through IT governance: a PLS-SEM analysis in Moroccan higher education institutions

    by Abdelilah Chahid, Souad Ahriz, Kamal El Guemmat, Khalifa Mansouri

    This study examines the impact of information technology governance (ITG) on digital transformation (DT) in Moroccan higher education institutions, particularly emphasising the mediating role of absorptive capacity. Utilising a rigorous methodological framework, the research analyzes data collected from 110 staff members using structural equation modelling with the SmartPLS tool. The goal is to explore the complex dynamics between ITG practices and DT capability. The findings reveal a positive and statistically significant relationship between ITG mechanisms and absorptive capacity (AC) and between the latter and the success of DT. The study also identifies AC as a crucial mediator between ITG and digital capability (DC). It suggests universities should strengthen their AC and adopt open policies to increase their innovative potential. This contribution enriches the existing literature by empirically confirming the influence of certain IT governance variables on DC within Moroccan universities, offering valuable insights for academic researchers and practitioners involved in IT governance strategies and DT.


  • Open Access

    Strengthening resilience against cyberattacks in Moroccan Universities through AHP, TOPSIS, and ITIL v4

    by Abdelilah Chahid, Souad Ahriz, Kamal El Guemmat, Khalifa Mansouri

    This study addresses the complex challenges of digital transformation in higher education by enhancing IT governance to combat cyber threats in Moroccan universities. By adopting a hybrid multi-criteria decision-making (MCDM) framework, the research combines the analytic hierarchy process (AHP) and the TOPSIS method to evaluate fourteen IT governance criteria, categorized into structural, procedural, and relational dimensions. Using TOPSIS, the study identifies the most relevant SVC services from the ITIL v4 value chain for each category, with the aim of developing an optimized strategic approach against cyberattacks. The input from ten academic experts was crucial in prioritizing these services. The results show that SVC services A5 and A2 are fundamental for optimizing the resources of structural and procedural mechanisms, while A4 and A2 play a key role in relational mechanisms. This strategic alignment enhances the resilience of Moroccan universities to cyber threats by ensuring a more efficient allocation of security resources and providing a robust defense against potential attacks.