About
As a Ph.D. in Computer Science, I specialize in complex network analysis and community detection, leveraging advanced computational techniques to uncover hidden patterns in social, biological, and technological systems. My research bridges theoretical innovation and practical applications, combining:
- Artificial Intelligence & Machine Learning (including deep learning and graph neural networks)
- Knowledge Graphs & Data Mining for structured knowledge extraction
- Multi-Agent Systems & Fuzzy Logic to model uncertainty in networked environments
Research Focus:
- Complex Networks: Modeling interactions in social/media/biological systems.
- AI-Driven Analysis: GNNs, knowledge graphs, and agent-based simulations.
- Open-Source Tools: Creating reproducible frameworks for network science.
Academic Activity
Education & Training
-
Ph.D. 2019-2025
Ph.D. in Computer Science and Artificial Intelligence
Chouaib Doukkali University, Faculty of Science, El Jadida, Morocco
-
M.S. 2007-2009
Master in Networks and Systems
Cadi Ayyad University, Faculty of Science Semlalia, Marrakech, Morocco
-
B.Sc. (Licence) 2002-2006
Bachlor degree in Computer Science
Cadi Ayyad University, Faculty of Science Semlalia, Marrakech, Morocco
Services
- AI & Machine Learning Consulting: Designing and deploying customized AI and machine learning solutions to enhance operational efficiency, foster innovation, and support data-driven decision-making.
- Data Science & Analytics: Extracting meaningful insights through advanced analytics, predictive modeling, and intelligent decision-support tools tailored to business needs.
- Professional Training & Capacity Building: Delivering practical, hands-on training programs and workshops in AI, Data Science, Federated Learning, and ERP systems for individuals, teams, and institutions.
Research Summary
Conducting research in an academic environment offers a unique opportunity to tackle real-world challenges while envisioning the technologies of tomorrow. My current research is driven by a deep interest in developing intelligent, decentralized, and privacy-preserving systems through artificial intelligence (AI), machine learning (ML), and data mining. As digital data continues to grow exponentially across all domains, from business and education to healthcare and scientific discovery, the need for effective data-driven methods becomes ever more pressing. Despite remarkable progress over the past two decades in both algorithmic development and computational resources, significant challenges remain in terms of realism, robustness, and ethical use of AI technologies.
My Ph.D. research has focused on complex network analysis, with a particular emphasis on community detection in attributed and dynamic networks. I have explored hybrid approaches that combine association rule learning, fuzzy logic, and deep learning to enhance the interpretability and adaptability of clustering models in large-scale social networks. This line of work contributes to a better understanding of how latent structures emerge in social interactions and how intelligent systems can uncover hidden patterns in evolving environments.
Additionally, I am interested in the intersection of AI with ethical computing and responsible data science, advocating for models that are transparent, fair, and respectful of privacy constraints. My applied research efforts strive to bridge theoretical advancements with practical, real-world implementations.
Looking forward, I aim to expand my research into emerging areas of collaborative AI systems that enhance data utility taking into consideration the semantic information hidden into nodes and edges attributes. I am particularly excited about cross-disciplinary collaborations that bridge computer science with healthcare, economics, and the social sciences. My research agenda is built around three pillars: developing novel learning models, creating robust and adaptive algorithms, and conducting rigorous experimental validations. I intend to explore innovative AI solutions to improve quality, trust, and interpretability in data-driven systems.
Potential project areas include, but are not limited to:
- Artificial Intelligence
- Machine Learning
- Deep Learning techniques
- Fuzzy Systems | Knowledge Graph
- Data Mining | MCDM
Research Projects
-
The first International Workshop on Modeling and Analysis of Complex Networks with Applications (MACNA-22) was held on August 9-11, 2022 in Ontario, Canada, in conjunction with The 17th International Conference on Future Networks and Communications (FNC) and The 19th International Conference on Mobile Systems and Pervasive Computing (MobiSPC).
This Workshop aims at bringing together practitioners, researchers and scientists to present their experimental and theoretical results within the scope of MACNA-22.
Selected Publications
I collaborate with interdisciplinary researchers to advance computational methods in data mining, complex systems, and network science, with a focus on community detection and multi-criteria decision analysis.
My work bridges theoretical innovation and practical applications, leveraging techniques like association rule learning, fuzzy logic, and deep learning to uncover patterns in social networks and beyond.
Filter by type:
Sort by year:
A comprehensive literature review on community detection: Approaches and applications.
Conference The 10th International Conference on Ambient Systems, Networks and Technologies (ANT 2019), April 29 - May 2, 2019. Leuven, Belgium.
Abstract
Community detection has been designed as an axial field in Complex Network Analysis (CNA), since it allows to reveal cohesive and meaningful sub-graphs, recognize the features, functions, structure and dynamic of such complex networks. In this sense, various methods and approaches have been developed over the years to provide appropriate solutions to complex network paradigms, especially to community detection problems. Meanwhile, identifying communities in a given complex network is a big challenge for scientists, which needs a significant amount of literature and survey. In this paper, we detail literatures on community detection for complex networks, because of the need for researchers to perform reviews on the main papers related to identification of communities in complex networks, and in order to point out their principal strengths and limitations. We believe that this literature contribution can be a valuable source of information in particular for practitioners in the field of community detection, and do not include all existing contributions. Therefore, we have been interested in the value of the contribution of the selected approaches more than the chronological order of the publications.
A Novel Approach of Community Detection Using Association Rules Learning: Application to User’s Friendships of Online Social Networks.
Conference the 5th International Conference on Smart City Applications, 7 - 9 October, 2020. Safranbolu, Türkiye.
Abstract
Both Social Network Analysis (SNA) and Association Rules Learning (ARL) enriched our daily-lives through various applications, by improving axial roles in several domains. In particular, the community detection in online social networks (OSN) has interested researchers, for its valuable contribution in understanding systems complexity, as either for academic, commercial or further purposes. The aim of this paper is the identification of communities in OSN using knowledge extraction based on association rules methods. Furthermore, we propose a new approach, namely ARL Clustering, using association rules learning for SNA. Particularly, we base our detection on user’s friendships of OSN by processing a four level technique to extract meaningful rules, converted later to communities. The conducted experimentation was applied on two synthetic real-world networks, and improved important results in identifying potential communities in comparison with existing approaches.
A k-Mean Classification Study of Eight Community Detection Algorithms: Application to Synthetic Social Network Datasets.
Book Chapter AI and IoT for Sustainable Development in Emerging Countries.
Abstract
An incredible importance has been devoted to the community detection algorithms applied to complex networks. In fact, a large variety of approaches and proposals have enriched both academic and commercial purposes to deal with the identification of communities and structures in many different fields such as sociology, biology, transportation, statistical physics, computer science and so on. In this paper, we propose a k-mean classification study of eight algorithms (Fast Greedy, Walktrap, Spinglass, Leading Eigen, Label propagation, Infomap, Optimal and Louvain), applied to two social network synthetic datasets. The aim of this paper is to highlight both convergences and divergences between them, taking into consideration the twofold: Modularity measure (Q) and the number of detected communities (CN). The experimentation has been fulfilled using the R language without focus on time and space complexity measurements. This contribution builds an experimental state of the art, designed to reach beginners audiences on the topic of community detection identification.
A multi-agent-based approach for community detection using association rules.
Journal International Journal of Data Science and Analytics
Abstract
In recent years, community detection has emerged as an important field of research, exerting a profound influence on various domains such as Social networks, Recommender systems, Citation networks, and Enterprise network. Acknowledging the profound implications of this development, we introduce a novel approach integrating a data mining technique, leveraging topical attributes of a network’s components. This approach seamlessly integrates with Social Network Analysis, within a multi-agent architecture composing four distinct hierarchical levels. In this paper, we present a novel approach for community detection that redirects the focus from traditional topological properties to topical properties of nodes and edges. This topical analysis perspective, often-neglected, constitutes the core of the proposed three-step methodology. We leverage the power of association rule mining using the Apriori algorithm as the initial step, extracting valuable insights from the network. Subsequently, we meticulously select meaningful rules, preparing them for the final stage where the proposed algorithm execution identifies both overlapped and non-overlapped communities within the network. To evaluate the effectiveness of our multi-agent system approach, we conducted tests on several real-world social networks, and performed comparisons with six traditional methods, thereby confirming the foundations of our approach.
ARLClustering: R Package for Community Detection-Based Association Rules Learning.
Conference International Conference on Engineering and Emerging Technologies (ICEET), 27-28 December 2024, Dubai, United Arab Emirates
Abstract
Community detection in social networks is a cornerstone of social network analysis, yet traditional methods often overlook the intricate interaction patterns between nodes, focusing instead on topological measures such as modularity or centrality. To address these limitations, this paper presents ARLClustering, an R package that employs association rule mining (ARM) through the Apriori algorithm to identify communities based on frequent interaction patterns. By leveraging ARM, ARLClustering uncovers nuanced relational structures, enabling the detection of meaningful and interpretable communities in complex networks. The used methodology involves transforming network data into a transactional format suitable for ARM, identifying frequent itemsets, and extracting communities from association rules. This approach is validated across diverse datasets (e.g. Karate Club, Dolphins, and Facebook networks and so on). A Comparative analysis highlights ARLClustering's important performance in detecting granular clusters and managing high volumes of network data while maintaining computational efficiency. Key contributions include its open-source accessibility, user-friendly design, and robust handling of parameter sensitivity for support and confidence thresholds. Beyond theoretical advancements, ARLClustering provides a practical R package for researchers, with comprehensive documentation and visualization features enabling comparative insights with traditional methods. Future work will enhance its visualization techniques, refine cluster evaluation metrics, and expand real-world applications to broaden its impact. Published on CRAN and available on GitHub, ARLClustering stands as a significant step forward in association rule-based community detection, bridging the gap between traditional methods and the demand for deeper relational insights in social network analysis.
Arlclustering: an R package for community detection in social networks based on user interaction and association rule learning.
Journal Applied Network Science (ANS)
Abstract
ARLClustering is an open-source R package for community detection in social networks. Unlike traditional methods that rely on structural properties such as modularity, degree centrality, and clustering coefficient, ARLClustering leverages association rule mining (ARM) to identify meaningful interaction patterns based on users’ friendship activity. By analyzing frequent interaction rules, it uncovers communities that may be overlooked by purely structural approaches. The package offers a comprehensive set of functions tailored for social network analysis. It has been tested on real-world datasets, including the Karate Club, Dolphins, LesMiserables, NetScience, and Facebook networks. The results demonstrate its effectiveness in detecting communities and provide a comparative analysis against existing methods. ARLClustering is now available on the CRAN repository, with its source code accessible on GitHub. It serves as a valuable tool for researchers and practitioners, not only enhancing community detection techniques in social network analysis but also introducing a novel approach to uncover hidden communities through experimental studies on real-world social network datasets.
A Fuzzy Logic-Based Approach for Community Detection in Social Networks: Analyzing Node Semantic Properties.
Journal International Journal of Fuzzy Systems (IJFS)
Abstract
Community detection in social networks constitutes an important field of study within the complex network analysis. It provides invaluable insights into the structures of complex networks, which are becoming increasingly prevalent in the current context of our interconnected world. Numerous community detection approaches proposed to lean on topological metrics, using centrality measures, clustering coefficient, and modularity. These metrics have been widely employed across diverse approaches. However, the question of how semantic properties as topical measure can enrich our understanding of the community structures remains an open research question. In this paper, we present a novel community detection approach leveraging node semantic properties within networks. In fact, we first employ the fuzzy logic to model the inherent uncertainty associated with the semantic attributes. Subsequently, we apply the k-means clustering algorithm to partition the network into communities based on the fuzzy membership function of different attributes. Our proposed approach initially recognizes the semantic dimensions within the node properties and establishes both the related decision and appreciation matrices. Then, it processes a quantitative transformation through the application of the proposed fuzzy technique. This transformation process yields a utility metric, which is later inputted into the k-means clustering algorithm to reveal the community structures present within the network. The proposed approach has been evaluated through experiments, conducted on various network sizes. The obtained results revealed significant findings, highlighting the effectiveness of the proposed approach.
A systematic review of deep learning methods for community detection in social networks.
Journal Frontiers in Artificial Intelligence.
Abstract
The rapid expansion of generated data through social networks has introduced significant challenges, which underscores the need for advanced methods to analyze and interpret these complex systems. Deep learning has emerged as an effective approach, offering robust capabilities to process large datasets, and uncover intricate relationships and patterns. In this systematic literature review, we explore research conducted over the past decade, focusing on the use of deep learning techniques for community detection in social networks. A total of 19 studies were carefully selected from reputable databases, including the ACM Library, Springer Link, Scopus, Science Direct, and IEEE Xplore. This review investigates the employed methodologies, evaluates their effectiveness, and discusses the challenges identified in these works. Our review shows that models like graph neural networks (GNNs), autoencoders, and convolutional neural networks (CNNs) are some of the most commonly used approaches for community detection. It also examines the variety of social networks, datasets, evaluation metrics, and employed frameworks in these studies. However, the analysis highlights several challenges, such as scalability, understanding how the models work (interpretability), and the need for solutions that can adapt to different types of networks. These issues stand out as important areas that need further attention and deeper research. This review provides meaningful insights for researchers working in social network analysis. It offers a detailed summary of recent developments, showcases the most impactful deep learning methods, and identifies key challenges that remain to be explored..
Semantic-Driven Community Detection in Complex Networks Using Fuzzy Logic and Multi-Criteria Decision Making.
Book Chapter AI-Driven Cybersecurity for Autonomous Systems. IGI Global
Abstract
This paper presents a novel approach to community detection in complex net- works by integrating fuzzy logic and multi-criteria decision-making techniques. Unlike traditional methods that rely primarily on topological metrics such as centrality, clustering coefficients, and modularity, our approach leverages the semantic attributes of nodes to uncover more meaningful community structures. By utilizing fuzzy logic, we address the inherent uncertainty and ambiguity in semantic attributes, providing a flexible framework for detecting communities that are not solely dependent on network topology. Furthermore, we enhance the scalability of our method by customizing the k-means clustering algorithm to accommodate networks of various sizes, from small-scale to large-scale net- works. Experimental results demonstrate that our fuzzy logic-based approach outperforms well-established algorithms such as K-means, Label Propagation, Louvain, and Leiden in the number of identified communities, especially in larger networks. The proposed method also shows robustness by generating balanced communities with competitive execution times, highlighting its computational feasibility for real-world network analysis. These findings suggest that our fuzzy logic approach offers a robust, scalable, and effective alternative to traditional community detection methods, providing new insights into the role of semantic attributes in network analysis and offering promising potential for a wide range of applications.
Contact | Meet Me
I'm always excited to connect with fellow researchers, potential collaborators, and students interested in network science and AI.
Whether you'd like to discuss my work, explore research opportunities, or simply chat about complex systems and community detection, feel free to reach out via email or connect on professional networks.
I'm particularly open to interdisciplinary projects bridging theory and real-world applications. Let's exchange ideas over coffee (virtual or in-person)
- Phone: +212 0615-412852
- el-moussaoui.m@ucd.ac.ma
- med.elmoussaoui.ced@gmail.com
- #assuom44
- Med El-Moussaoui, Ph.D.
Gallery →
No news!