Abdelhak Boulaalam
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Abdelhak Boulaalam has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- El Habib Nfaoui
Sidi Mohamed Ben Abdellah University
10 shared publications - Abderrahim Zannou
ERCI2A, Faculty of Sciences and Techniques, Al Hoceima, Abdelmalek Essaadi University, Tetouan 93000, Morocco
9 shared publications El Habib Nfaoui
University Sidi Mohamed Ben Abdellah
5 shared publications- Fahd Kalloubi
Mohammed VI Polytechnic University UM6P , Vanguard Center , Benguerir , Morocco
4 shared publications - Soukaina Bouarourou
Computer Science Department, Faculty of Sciences, Mohamed V University in Rabat, Rabat, Morocco
4 shared publications - Youssef Aounzou
LSI Laboratory, National Schools of Applied Sciences , Sidi Mohamed Ben Abdellah University , Fez , Morocco
4 shared publications
Latest Publications
- Open Access
- 23 September 2026
- Published by Elsevier in SSRN Electronic Journal
SPCA: SubPath Semantic Encoding for Context Aware HeterogeneousGraph Neural Networks
Heterogeneous graph neural networks often rely on metapaths to guide message passing and capture composite relations. We argue that the intermediate node-type sequence along a metapath, the subpath, carries distinct semantic signal that is underused when metapaths are treated as single units. We propose SPCA (SubPath Context-Aware) convolution, a layer design that couples metapath-based neighbor aggregation with an explicit subpath context encoder, so that node representations are informed by both structural neighborhoods and the schema-level "path type" defined by intermediate types. Subpath embeddings are computed from learnable shared type embeddings and combined with structural features via a simpleaggregation mechanism and layer normalization. We evaluate SPCA on node classification, clustering, and link prediction over standard benchmarks. Results show consistent gains over strong baselines, with ablations confirming the role of subpath encoding and a complexity profile between highly expressive and highly efficient prior art.
- 23 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessSPCA: SubPath Semantic Encoding for Context Aware HeterogeneousGraph Neural Networks
Heterogeneous graph neural networks often rely on metapaths to guide message passing and capture composite relations. We argue that the intermediate node-type sequence along a metapath, the subpath, carries distinct semantic signal that is underused when metapaths are treated as single units. We propose SPCA (SubPath Context-Aware) convolution, a layer design that couples metapath-based neighbor aggregation with an explicit subpath context encoder, so that node representations are informed by both structural neighborhoods and the schema-level "path type" defined by intermediate types. Subpath embeddings are computed from learnable shared type embeddings and combined with structural features via a simpleaggregation mechanism and layer normalization. We evaluate SPCA on node classification, clustering, and link prediction over standard benchmarks. Results show consistent gains over strong baselines, with ablations confirming the role of subpath encoding and a complexity profile between highly expressive and highly efficient prior art.
- 29 May 2026
- Published by Springer Nature
- p. 111-121
Lung Cancer Classification Using Deep Learning
- Open Access
- 16 May 2026
- Published by Elsevier in SSRN Electronic Journal
TAHAN: Type-Aware Heterogeneous Graph AttentionNeural Networks
Learning from heterogeneous graphs remains challenging when structural complexity surpasses the modeling capacity of existing heterogeneous graph neural networks (HGNNs). Although current HGNNs capture structural and semantic relations, they typically treat node types as static categorical labels and overlook the contextual influence these types exert during message passing. Metapath-based models further restrict aggregation to the endpoints of a metapath, neglecting the semantic contribution of intermediate node types. As a result, they underexploit type-dependent contextual signals and often struggle in settings with sparse or weakly connected entity types. We propose TAHAN (Type-Aware Heterogeneous Graph Attention Network), a model that explicitly incorporates type-sensitive and context-dependent information into heterogeneous neighborhood aggregation. TAHAN first gathers neighbors across full metapath instances and multiple neighborhood levels, organizing them into type-specific neighborhoods. It then applies a two-stage attention mechanism: type-specific attention to capture contributions from nodes within the same type, followed by type-aware attention to model interactions across different types. The resulting representations fuse these signals into expressive, type-sensitive node embeddings. Experiments on multiple benchmarks show that TAHAN consistently outperforms state-of-the-art HGNNs. Component-level analyses further demonstrate the interpretability and practical value of type-aware modeling, offering insights for future research.
- 16 May 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessTAHAN: Type-Aware Heterogeneous Graph AttentionNeural Networks
Learning from heterogeneous graphs remains challenging when structural complexity surpasses the modeling capacity of existing heterogeneous graph neural networks (HGNNs). Although current HGNNs capture structural and semantic relations, they typically treat node types as static categorical labels and overlook the contextual influence these types exert during message passing. Metapath-based models further restrict aggregation to the endpoints of a metapath, neglecting the semantic contribution of intermediate node types. As a result, they underexploit type-dependent contextual signals and often struggle in settings with sparse or weakly connected entity types. We propose TAHAN (Type-Aware Heterogeneous Graph Attention Network), a model that explicitly incorporates type-sensitive and context-dependent information into heterogeneous neighborhood aggregation. TAHAN first gathers neighbors across full metapath instances and multiple neighborhood levels, organizing them into type-specific neighborhoods. It then applies a two-stage attention mechanism: type-specific attention to capture contributions from nodes within the same type, followed by type-aware attention to model interactions across different types. The resulting representations fuse these signals into expressive, type-sensitive node embeddings. Experiments on multiple benchmarks show that TAHAN consistently outperforms state-of-the-art HGNNs. Component-level analyses further demonstrate the interpretability and practical value of type-aware modeling, offering insights for future research.
- Open Access
- 25 March 2025
- Published by MDPI AG in Future Internet
- Vol. 17 (4) , 140
Enhancing IoT Scalability and Interoperability Through Ontology Alignment and FedProx
The rapid expansion of IoT devices has introduced major challenges in ensuring data interoperability, enabling real-time processing, and achieving scalability, especially in decentralized edge computing environments. In this paper, an advanced framework of FedProx with ontology-driven data standardization is proposed, which can meet such challenges comprehensively. On the one hand, it can guarantee semantic consistency across different kinds of IoT devices using unified ontology, so that data from multiple sources could be seamlessly integrated; on the other hand, it solves the non-IID issues of data and limited resources in edge servers by FedProx. Experimental findings indicate that FedProx outperforms FedAvg, with a remarkable accuracy level of 89.4%, having higher convergence rates, and attaining a 30% saving on communication overhead through gradient compression. In addition, the ontology alignment procedure yielded a 95% success rate, thereby ensuring uniform data preprocessing across domains, including traffic monitoring and parking management. The model demonstrates outstanding scalability and flexibility to new devices, while maintaining high performance during ontology evolution. These findings highlight its great potential for deployment in smart cities, environmental monitoring, and other IoT-based ecosystems, thereby enabling the creation of more efficient and integrated solutions in these areas.
- 25 March 2025
- Published by MDPI AG in Future Internet
- Vol. 17 (4) , 140
Open AccessEnhancing IoT Scalability and Interoperability Through Ontology Alignment and FedProx
The rapid expansion of IoT devices has introduced major challenges in ensuring data interoperability, enabling real-time processing, and achieving scalability, especially in decentralized edge computing environments. In this paper, an advanced framework of FedProx with ontology-driven data standardization is proposed, which can meet such challenges comprehensively. On the one hand, it can guarantee semantic consistency across different kinds of IoT devices using unified ontology, so that data from multiple sources could be seamlessly integrated; on the other hand, it solves the non-IID issues of data and limited resources in edge servers by FedProx. Experimental findings indicate that FedProx outperforms FedAvg, with a remarkable accuracy level of 89.4%, having higher convergence rates, and attaining a 30% saving on communication overhead through gradient compression. In addition, the ontology alignment procedure yielded a 95% success rate, thereby ensuring uniform data preprocessing across domains, including traffic monitoring and parking management. The model demonstrates outstanding scalability and flexibility to new devices, while maintaining high performance during ontology evolution. These findings highlight its great potential for deployment in smart cities, environmental monitoring, and other IoT-based ecosystems, thereby enabling the creation of more efficient and integrated solutions in these areas.
- Open Access
- 12 February 2025
- Published by International Association of Online Engineering (IAOE) in International Journal of Interactive Mobile Technologies (iJIM)
- Vol. 19 (03) , 209-226
IoT Challenges and Issues: A Comprehensive Review of Software Defined Networking and Network Function Virtualization Solutions
The idea behind the Internet of Things (IoT) is to connect everything, including laptops, smartphones, sensors, and other devices, to the Internet. To build an autonomous environment without human intervention. This novel network was used in several industries, including smart homes, smart cities, healthcare, etc. For this reason, IoT networks are growing in infrastructure. As a result, the administration of this vast array of linked devices and produced data becomes more complicated. Thus, a new elastic mechanism is required for this dynamic and rapid evolution in configuration, control, management, etc. Network Function Virtualization (NFV) and Software Defined Networking (SDN) have become essential points in scientific research to overcome IoT challenges such as security, heterogeneity, energy efficiency, interoperability, and more. These two approaches have proven their efficiency in adapting to dynamic and evolving networks. SDN reduces network latency by up to 30% and increases device scalability by 40%. At the same time, NFV optimizes resource allocation, achieving up to a 35% reduction in energy consumption and a 20% decrease in operational costs through virtualized infrastructure. In this review, we systematically analyze solutions designed for IoT systems by developing a state-of-the-art for NFV and SDN and thoroughly researching the various problems that IoT will face. Thus, we compare SDN- and NFV-based IoT solutions to overcome these challenges. Lastly, we will discuss the different obstacles that can lower the performance of SDN/NFV applications on the IoT. The contribution of this review lies in its systematic evaluation and comparison of current NFV and SDN approaches, providing valuable insights and paving the way for future research to enhance the integration and management of IoT systems.
- 12 February 2025
- Published by International Association of Online Engineering (IAOE) in International Journal of Interactive Mobile Technologies (iJIM)
- Vol. 19 (03) , 209-226
Open AccessIoT Challenges and Issues: A Comprehensive Review of Software Defined Networking and Network Function Virtualization Solutions
The idea behind the Internet of Things (IoT) is to connect everything, including laptops, smartphones, sensors, and other devices, to the Internet. To build an autonomous environment without human intervention. This novel network was used in several industries, including smart homes, smart cities, healthcare, etc. For this reason, IoT networks are growing in infrastructure. As a result, the administration of this vast array of linked devices and produced data becomes more complicated. Thus, a new elastic mechanism is required for this dynamic and rapid evolution in configuration, control, management, etc. Network Function Virtualization (NFV) and Software Defined Networking (SDN) have become essential points in scientific research to overcome IoT challenges such as security, heterogeneity, energy efficiency, interoperability, and more. These two approaches have proven their efficiency in adapting to dynamic and evolving networks. SDN reduces network latency by up to 30% and increases device scalability by 40%. At the same time, NFV optimizes resource allocation, achieving up to a 35% reduction in energy consumption and a 20% decrease in operational costs through virtualized infrastructure. In this review, we systematically analyze solutions designed for IoT systems by developing a state-of-the-art for NFV and SDN and thoroughly researching the various problems that IoT will face. Thus, we compare SDN- and NFV-based IoT solutions to overcome these challenges. Lastly, we will discuss the different obstacles that can lower the performance of SDN/NFV applications on the IoT. The contribution of this review lies in its systematic evaluation and comparison of current NFV and SDN approaches, providing valuable insights and paving the way for future research to enhance the integration and management of IoT systems.
- Open Access
- 1 January 2025
- Published by Walter de Gruyter GmbH in International Journal on Smart Sensing and Intelligent Systems
- Vol. 18 (1)
Convergence of blockchain, IoT, and machine learning: exploring opportunities and challenges – a systematic review
With the rapid technological advancement, the convergence of the Internet of Things (IoT), blockchain, and machine learning (ML) has a significant revolution across industries, driving them in areas like Industry 4.0, smart cities, and advanced transportation systems. This study presents a comprehensive review of the integration of the IoT, blockchain, and ML technologies, highlighting their transformative impact on diverse industries. It addresses the challenges of scalability, security, and data management posed by the growth of interconnected IoT devices, proposing solutions through advanced algorithms and the integration of blockchain for data security and immutability. By examining real-world applications and reviewing existing literature, this study underscores the potential benefits and complexities of merging these technologies, while also identifying research challenges and future directions. Overall, it offers insights into the innovative constructive collaboration of IoT, blockchain, and ML, which has become the most important of modern technological advances.
- 1 January 2025
- Published by Walter de Gruyter GmbH in International Journal on Smart Sensing and Intelligent Systems
- Vol. 18 (1)
Open AccessConvergence of blockchain, IoT, and machine learning: exploring opportunities and challenges – a systematic review
With the rapid technological advancement, the convergence of the Internet of Things (IoT), blockchain, and machine learning (ML) has a significant revolution across industries, driving them in areas like Industry 4.0, smart cities, and advanced transportation systems. This study presents a comprehensive review of the integration of the IoT, blockchain, and ML technologies, highlighting their transformative impact on diverse industries. It addresses the challenges of scalability, security, and data management posed by the growth of interconnected IoT devices, proposing solutions through advanced algorithms and the integration of blockchain for data security and immutability. By examining real-world applications and reviewing existing literature, this study underscores the potential benefits and complexities of merging these technologies, while also identifying research challenges and future directions. Overall, it offers insights into the innovative constructive collaboration of IoT, blockchain, and ML, which has become the most important of modern technological advances.
- Open Access
A Blockchain of Things System for Managing Handcrafted Products in a Cultural Industry
The handicraft sector is often seen as a symbol of a country’s cultural identity, as it relies on specialized traditional techniques, skills, and knowledge that are often passed down through generations. As a result, ensuring the authenticity and integrity of this creative value has become a significant challenge, especially with the growth of counterfeiting techniques in this industry. Thus, integrating digital technologies into such sectors offers numerous operational benefits such as transparency, decentralization, data security, and authenticity needs. This paper presents an innovative approach for the handicraft sector, which exploits blockchain technology and the Internet of Things to guarantee the authenticity of cultural heritage. Through experimental evaluations comparing the decentralized blockchain-based system with traditional centralized methods using key metrics such as response time and transactions per second, this study reveals significant results. The statistical analysis reveals that the decentralized approach improves performance in terms of response times for verification and addition processes compared to the centralized system. Specifically, verification is approximately 4.66 times faster and addition is approximately 4.30 times faster in a decentralized system. However, transaction latency in the decentralized approach is approximately 38.21% higher than in the centralized system.
Open AccessA Blockchain of Things System for Managing Handcrafted Products in a Cultural Industry
The handicraft sector is often seen as a symbol of a country’s cultural identity, as it relies on specialized traditional techniques, skills, and knowledge that are often passed down through generations. As a result, ensuring the authenticity and integrity of this creative value has become a significant challenge, especially with the growth of counterfeiting techniques in this industry. Thus, integrating digital technologies into such sectors offers numerous operational benefits such as transparency, decentralization, data security, and authenticity needs. This paper presents an innovative approach for the handicraft sector, which exploits blockchain technology and the Internet of Things to guarantee the authenticity of cultural heritage. Through experimental evaluations comparing the decentralized blockchain-based system with traditional centralized methods using key metrics such as response time and transactions per second, this study reveals significant results. The statistical analysis reveals that the decentralized approach improves performance in terms of response times for verification and addition processes compared to the centralized system. Specifically, verification is approximately 4.66 times faster and addition is approximately 4.30 times faster in a decentralized system. However, transaction latency in the decentralized approach is approximately 38.21% higher than in the centralized system.
- 1 January 2024
- Published by Springer Nature
- p. 32-47
Neural Microblogging Followee Recommender System Based on Pre-trained Transformer, and Topic Models
- 1 January 2024
- Published by Springer Nature
- p. 32-47
Neural Microblogging Followee Recommender System Based on Pre-trained Transformer, and Topic Models
- 1 January 2024
- Published by Springer Nature
- p. 175-186
Handicraft Supply Chain Management: Towards an Intelligent System Based on IoT, Blockchain, and Machine Learning
- 1 January 2024
- Published by Springer Nature
- p. 175-186
Handicraft Supply Chain Management: Towards an Intelligent System Based on IoT, Blockchain, and Machine Learning
- 1 January 2024
- Published by Springer Nature
- p. 913-923
IoT-Powered Big Data Processing for Intelligent Transportation: Enhancing Effectiveness and Forecasting
- 1 January 2024
- Published by Springer Nature
- p. 913-923
IoT-Powered Big Data Processing for Intelligent Transportation: Enhancing Effectiveness and Forecasting