Habiba Drias
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Habiba Drias has a total of 180 co-authors. Here are the top co-authors by number of shared publications:
- Ilyes Khennak
USTHB
21 shared publications - Yassine Drias
MEF University
17 shared publications - Dalila Boughaci
University of Science and Technology Houari Boumediene
13 shared publications - Naila Aziza Houacine
LRIA, USTHB
10 shared publications - Hadjer Moulai
USTHB
8 shared publications Nadjet Kamel
Ferhat Abbas University Sétif Algeria
7 shared publications
Latest Publications
- 31 October 2026
- Published by Turkish Academy of Sciences
- p. 215-234
Artificial Intelligence Governance for Public Administration and Technology Development and Use
Artificial intelligence (AI), once a niche research area, has become a powerful and versatile technology shaping economies, public institutions, and everyday life. As AI systems become more powerful and more deeply integrated into critical sectors, from healthcare and education to finance and national security, the question of their governance becomes crucial. This article examines the main challenges of AI governance. Existing regulatory frameworks and approaches, as well as effective AI governance, are analyzed, along with the governance of cutting-edge and general-purpose AI models. The fundamental dilemma of AI governance lies in finding a balance between innovation, security, and societal values. This point is highlighted and the emergence of AI governance as a crucial global issue is explored. As illustrative examples, the governance for municipal administration is addressed and a case study on predicting climate change demonstrates how AI can contribute to better governance. The spatial data used, as well as the algorithmic model are presented and the potential risks and disruptions that may arise are discussed for both cases.
- 31 October 2026
- Published by Turkish Academy of Sciences
- p. 215-234
Artificial Intelligence Governance for Public Administration and Technology Development and Use
Artificial intelligence (AI), once a niche research area, has become a powerful and versatile technology shaping economies, public institutions, and everyday life. As AI systems become more powerful and more deeply integrated into critical sectors, from healthcare and education to finance and national security, the question of their governance becomes crucial. This article examines the main challenges of AI governance. Existing regulatory frameworks and approaches, as well as effective AI governance, are analyzed, along with the governance of cutting-edge and general-purpose AI models. The fundamental dilemma of AI governance lies in finding a balance between innovation, security, and societal values. This point is highlighted and the emergence of AI governance as a crucial global issue is explored. As illustrative examples, the governance for municipal administration is addressed and a case study on predicting climate change demonstrates how AI can contribute to better governance. The spatial data used, as well as the algorithmic model are presented and the potential risks and disruptions that may arise are discussed for both cases.
- 30 September 2026
- Published by Wiley in Expert Systems
- Vol. 43 (11) , e70438
A Stochastic Agnostic Meta‐Clustering Framework: Application to African Agro‐Ecological Zoning
This article introduces Stochastic Agnostic Meta‐Clustering (SAMC), a novel ensemble clustering framework designed to address the challenges of high‐dimensional and uncertain data. SAMC integrates stochastic feature subspace decomposition with metaheuristic optimization to support both hard and soft clustering, making it well suited for domains characterized by gradual transitions and fuzzy boundaries, such as environmental and agricultural systems. The proposed framework is evaluated on an agro‐ecological zoning (AEZ) task using data structured around SCORPAN factors and is compared against established hard and soft clustering methods, with and without dimensionality reduction. Experimental results demonstrate the SAMC's advantageous trade‐off between interpretability and scalability while improving robustness to feature heterogeneity and preventing over‐fragmentation. In addition, post‐clustering interpretability analyses are conducted to identify dominant agro‐ecological drivers and representative rules characterizing the resulting zones. The comparative analysis further confirms that soft clustering approaches are better suited to capturing the fuzzy boundaries inherent in environmental systems and provided a soft clustering‐based AEZ analysis at the continental scale for Africa.
- 30 September 2026
- Published by Wiley in Expert Systems
- Vol. 43 (11) , e70438
A Stochastic Agnostic Meta‐Clustering Framework: Application to African Agro‐Ecological Zoning
This article introduces Stochastic Agnostic Meta‐Clustering (SAMC), a novel ensemble clustering framework designed to address the challenges of high‐dimensional and uncertain data. SAMC integrates stochastic feature subspace decomposition with metaheuristic optimization to support both hard and soft clustering, making it well suited for domains characterized by gradual transitions and fuzzy boundaries, such as environmental and agricultural systems. The proposed framework is evaluated on an agro‐ecological zoning (AEZ) task using data structured around SCORPAN factors and is compared against established hard and soft clustering methods, with and without dimensionality reduction. Experimental results demonstrate the SAMC's advantageous trade‐off between interpretability and scalability while improving robustness to feature heterogeneity and preventing over‐fragmentation. In addition, post‐clustering interpretability analyses are conducted to identify dominant agro‐ecological drivers and representative rules characterizing the resulting zones. The comparative analysis further confirms that soft clustering approaches are better suited to capturing the fuzzy boundaries inherent in environmental systems and provided a soft clustering‐based AEZ analysis at the continental scale for Africa.
- 8 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. PP (15249050) , 1-22
Deep Reinforcement Learning for Cooperative Intelligent Transportation Systems: A Survey on Architecture, Use Cases, and Future Directions
The emergence of Cooperative Intelligent Transportation Systems (C-ITS) has revolutionized urban mobility by enabling seamless collaboration among vehicles, infrastructure, and individuals to improve traffic management, safety, and efficiency. Deep Reinforcement Learning (DRL) has become a key technology in this ecosystem, empowering autonomous agents to make real-time decisions that optimize traffic flow, reduce congestion, and enhance road safety. Although many surveys on Intelligent Transportation Systems (ITS) either overlook cooperative aspects or primarily emphasize security, this paper bridges the gap by examining the diverse applications of DRL in C-ITS. It examines critical areas such as traffic signal control, AV coordination, route planning, and human-vehicle interaction. The study also traces the evolution of DRL algorithms, their adaptation to transportation challenges, and their integration with cutting-edge projects and standards. Additionally, the paper provides a comprehensive analysis of current research trends, identifying achievements, unresolved challenges, and future directions in the field. By synthesizing existing literature and highlighting the synergy between DRL and C-ITS, this survey serves as a valuable resource for researchers, policymakers, and industry professionals striving to develop intelligent, cooperative, and sustainable transportation systems. The insights offered aim to guide advancements in this rapidly growing domain, fostering innovation and practical implementation.
- 8 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. PP (15249050) , 1-22
Deep Reinforcement Learning for Cooperative Intelligent Transportation Systems: A Survey on Architecture, Use Cases, and Future Directions
The emergence of Cooperative Intelligent Transportation Systems (C-ITS) has revolutionized urban mobility by enabling seamless collaboration among vehicles, infrastructure, and individuals to improve traffic management, safety, and efficiency. Deep Reinforcement Learning (DRL) has become a key technology in this ecosystem, empowering autonomous agents to make real-time decisions that optimize traffic flow, reduce congestion, and enhance road safety. Although many surveys on Intelligent Transportation Systems (ITS) either overlook cooperative aspects or primarily emphasize security, this paper bridges the gap by examining the diverse applications of DRL in C-ITS. It examines critical areas such as traffic signal control, AV coordination, route planning, and human-vehicle interaction. The study also traces the evolution of DRL algorithms, their adaptation to transportation challenges, and their integration with cutting-edge projects and standards. Additionally, the paper provides a comprehensive analysis of current research trends, identifying achievements, unresolved challenges, and future directions in the field. By synthesizing existing literature and highlighting the synergy between DRL and C-ITS, this survey serves as a valuable resource for researchers, policymakers, and industry professionals striving to develop intelligent, cooperative, and sustainable transportation systems. The insights offered aim to guide advancements in this rapidly growing domain, fostering innovation and practical implementation.
- 25 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 23 (15455955) , 7000-7012
Optimizing Timetable Scheduling: A Smart Local Search Approach With Aspiration and Random Moves Strategies
This paper presents the Smart Local Search (SLS) algorithm, a hybrid metaheuristic framework for the NP-hard examination timetabling problem. SLS distinctively integrates graph-based construction heuristics with a guided local search framework, enhanced by an aspiration criterion to recover promising solutions overlooked by penalty terms and a strategic random move mechanism to escape local optima. Extensive experiments on all 11 of Carter’s widely used benchmark datasets show that SLS achieves competitive results, with an average penalty of 28.03 and a standard deviation of 42.30 across the benchmarks. Statistical significance testing (Wilcoxon signed-rank test, α = 0.05) confirms that SLS’s performance is significantly better than most baseline methods (p-value < 0.05). This conclusion is further supported by highly significant results against the average of all methods (p-value = 0.001953) and in an aggregate comparison with all competing results (p-value = 3.93×10−10). The algorithm demonstrates particular strength in stability and consistency, making it a robust and efficient solution for complex educational scheduling environments.
- 25 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. 23 (15455955) , 7000-7012
Optimizing Timetable Scheduling: A Smart Local Search Approach With Aspiration and Random Moves Strategies
This paper presents the Smart Local Search (SLS) algorithm, a hybrid metaheuristic framework for the NP-hard examination timetabling problem. SLS distinctively integrates graph-based construction heuristics with a guided local search framework, enhanced by an aspiration criterion to recover promising solutions overlooked by penalty terms and a strategic random move mechanism to escape local optima. Extensive experiments on all 11 of Carter’s widely used benchmark datasets show that SLS achieves competitive results, with an average penalty of 28.03 and a standard deviation of 42.30 across the benchmarks. Statistical significance testing (Wilcoxon signed-rank test, α = 0.05) confirms that SLS’s performance is significantly better than most baseline methods (p-value < 0.05). This conclusion is further supported by highly significant results against the average of all methods (p-value = 0.001953) and in an aggregate comparison with all competing results (p-value = 3.93×10−10). The algorithm demonstrates particular strength in stability and consistency, making it a robust and efficient solution for complex educational scheduling environments.
- Open Access
- 1 March 2026
- Published by EDP Sciences in RAIRO - Operations Research
- Vol. 60 (2) , 1025-1052
Cultural inheritance through Swarm intelligence learning
Cultural inheritance is the main factor for the evolution and skills development through generations. This paper exploits two Machine Learning techniques to imitate the learning process that constitutes the cultural inheritance of Elephants in a Swarm Intelligence (SI) algorithm called the Robotic-Elephant Herding Optimization (REHO). The first REHO-based learning utilizes a GPU k- Nearest Neighbors (GPU-kNN) algorithm, while the second is based on a Deep Neural Network. The performances of the Machine Learning (ML) algorithms are evaluated through five well-known metrics. Their effectiveness in solving the Target Detection Problem is compared with the original REHO via several experiments. The results show that the Deep Learning (DL) model performs better than the GPU-KNN method. These findings prompted us to propose a new hybrid SI and DL model algorithm, namely Robotic-Elephant and Particle Cultural Algorithm (REPCA). This novel approach combines the strength of the REHO-based learning model with the dynamic velocity accelerator of Particle Swarm Optimization (PSO) and knowledge inheritance provided by the Cultural Algorithm. The application problem consists of a swarm of robots that cooperate to detect an exponentially increasing number of targets, known as the Dynamic Target Detection Problem (DTDP). The proposed REPCA is compared with up-to-date algorithms on recent and real-world applications covering the robotics field and the Covid-19 pandemic. The results show a significant positive impact that the cultural inheritance and hybrid methods bring to solve the DTDP.
- 1 March 2026
- Published by EDP Sciences in RAIRO - Operations Research
- Vol. 60 (2) , 1025-1052
Open AccessCultural inheritance through Swarm intelligence learning
Cultural inheritance is the main factor for the evolution and skills development through generations. This paper exploits two Machine Learning techniques to imitate the learning process that constitutes the cultural inheritance of Elephants in a Swarm Intelligence (SI) algorithm called the Robotic-Elephant Herding Optimization (REHO). The first REHO-based learning utilizes a GPU k- Nearest Neighbors (GPU-kNN) algorithm, while the second is based on a Deep Neural Network. The performances of the Machine Learning (ML) algorithms are evaluated through five well-known metrics. Their effectiveness in solving the Target Detection Problem is compared with the original REHO via several experiments. The results show that the Deep Learning (DL) model performs better than the GPU-KNN method. These findings prompted us to propose a new hybrid SI and DL model algorithm, namely Robotic-Elephant and Particle Cultural Algorithm (REPCA). This novel approach combines the strength of the REHO-based learning model with the dynamic velocity accelerator of Particle Swarm Optimization (PSO) and knowledge inheritance provided by the Cultural Algorithm. The application problem consists of a swarm of robots that cooperate to detect an exponentially increasing number of targets, known as the Dynamic Target Detection Problem (DTDP). The proposed REPCA is compared with up-to-date algorithms on recent and real-world applications covering the robotics field and the Covid-19 pandemic. The results show a significant positive impact that the cultural inheritance and hybrid methods bring to solve the DTDP.
- Open Access
- 17 November 2025
- Published by EDP Sciences in RAIRO - Operations Research
- Vol. 59 (6) , 3455-3486
Slime Mould Algorithm-based mutated Chaotic Local Search for global optimization and engineering applications
Swarm intelligence has gained increasing interest, with the Slime Mould Algorithm (SMA) standing out as a promising yet convergence-prone metaheuristic. To address this limitation, we propose an improved variant, called mSMACLS, which integrates a mutated Chaotic Local Search (mCLS) strategy into SMA. The mCLS mechanism selectively perturbs dimensions of the global best solution using chaotic sequences, enhancing local exploitation while maintaining global exploration. We evaluate 10 chaotic maps to identify the most effective configuration and conduct extensive experiments on 23 standard benchmark functions and 10 more complex optimization problems. The proposed algorithm is further validated on 15 challenging benchmark problems from the CEC2017 competition on constrained single-objective optimization. Results show that the Logistic chaotic map provides the best performance within the mCLS framework. Overall, mSMACLS consistently outperforms six well-known metaheuristics across most test cases, demonstrating improved convergence speed and solution quality. Finally, the algorithm’s applicability is tested on 3 real-world engineering design problems, and the results were very encouraging. The source code is publicly available at https://www.mathworks.com/matlabcentral/fileexchange/182303-slime-mould-algorithm-based-mutated-chaotic-local-search .
- 17 November 2025
- Published by EDP Sciences in RAIRO - Operations Research
- Vol. 59 (6) , 3455-3486
Open AccessSlime Mould Algorithm-based mutated Chaotic Local Search for global optimization and engineering applications
Swarm intelligence has gained increasing interest, with the Slime Mould Algorithm (SMA) standing out as a promising yet convergence-prone metaheuristic. To address this limitation, we propose an improved variant, called mSMACLS, which integrates a mutated Chaotic Local Search (mCLS) strategy into SMA. The mCLS mechanism selectively perturbs dimensions of the global best solution using chaotic sequences, enhancing local exploitation while maintaining global exploration. We evaluate 10 chaotic maps to identify the most effective configuration and conduct extensive experiments on 23 standard benchmark functions and 10 more complex optimization problems. The proposed algorithm is further validated on 15 challenging benchmark problems from the CEC2017 competition on constrained single-objective optimization. Results show that the Logistic chaotic map provides the best performance within the mCLS framework. Overall, mSMACLS consistently outperforms six well-known metaheuristics across most test cases, demonstrating improved convergence speed and solution quality. Finally, the algorithm’s applicability is tested on 3 real-world engineering design problems, and the results were very encouraging. The source code is publicly available at https://www.mathworks.com/matlabcentral/fileexchange/182303-slime-mould-algorithm-based-mutated-chaotic-local-search .
- 30 October 2025
- Published by Springer Nature in Quantum Information Processing
- Vol. 24 (11) , 1-33
Methane emissions forecasting using hybrid quantum–classical deep learning models: case study of North Africa
- 30 October 2025
- Published by Springer Nature in Quantum Information Processing
- Vol. 24 (11) , 1-33
Methane emissions forecasting using hybrid quantum–classical deep learning models: case study of North Africa
- Open Access
- 29 May 2025
- Published by Cerebration Science Publishing Co., Limited in International Journal of Computer Information Systems and Industrial Management Applications
- Vol. 17, 387-407
GPU-Accelerated Slime Mould Algorithm for Urgent Transportation in Disaster Response: A COVID-19 Application
In the face of increasing natural disasters, ensuring rapid patient transportation by Emergency Medical Services (EMS) is critical to saving lives. The Ambulance Dispatching and Relocation Problem (ADRP) poses a significant challenge, requiring swift allocation of limited ambulance resources. To address this issue, we propose a GPU-Accelerated Slime Mould Algorithm (GPU-SMA) designed for real-time decision-making in disaster scenarios. Our approach leverages the parallel processing power of GPUs using the Compute Unified Device Architecture (CUDA). This significantly reduces computational time, enabling faster and more effective optimization. Additionally, we introduce a new relocation policy that utilizes real-time ambulance data to maintain optimal ambulance positioning. Our method has been using real-world data from the COVID-19 pandemic in Chicago. The results show a remarkable 23x speedup on GeForce RTX 3090 and RTX A4000 GPUs compared to the serial implementation. GPU-SMA outperforms five leading parallel algorithms (GPU-PSO, GPU-APSO, GPU-HHO, GPU-FA, and GPU-BA) in efficiency and effectiveness. A Friedman test confirms the statistical significance of these results.
- 29 May 2025
- Published by Cerebration Science Publishing Co., Limited in International Journal of Computer Information Systems and Industrial Management Applications
- Vol. 17, 387-407
Open AccessGPU-Accelerated Slime Mould Algorithm for Urgent Transportation in Disaster Response: A COVID-19 Application
In the face of increasing natural disasters, ensuring rapid patient transportation by Emergency Medical Services (EMS) is critical to saving lives. The Ambulance Dispatching and Relocation Problem (ADRP) poses a significant challenge, requiring swift allocation of limited ambulance resources. To address this issue, we propose a GPU-Accelerated Slime Mould Algorithm (GPU-SMA) designed for real-time decision-making in disaster scenarios. Our approach leverages the parallel processing power of GPUs using the Compute Unified Device Architecture (CUDA). This significantly reduces computational time, enabling faster and more effective optimization. Additionally, we introduce a new relocation policy that utilizes real-time ambulance data to maintain optimal ambulance positioning. Our method has been using real-world data from the COVID-19 pandemic in Chicago. The results show a remarkable 23x speedup on GeForce RTX 3090 and RTX A4000 GPUs compared to the serial implementation. GPU-SMA outperforms five leading parallel algorithms (GPU-PSO, GPU-APSO, GPU-HHO, GPU-FA, and GPU-BA) in efficiency and effectiveness. A Friedman test confirms the statistical significance of these results.
- 21 April 2025
- Published by Springer Nature
- p. 30-45
A Data-Driven Wildfire Forecasting: Case Study of Algeria
- 21 April 2025
- Published by Springer Nature
- p. 30-45
A Data-Driven Wildfire Forecasting: Case Study of Algeria