SAMIR Akhrouf

Amroune Houssam Eddine

A digital academy to learn programming, artificial intelligence and IT field

Loumi Mohamed Amine , Guerra Manar

A Mobile Application for Mapping Forest Fire Risk in Algeria using GIS-AHP Method

Zerig Salah Eddine , Arslane Zakaria, Dhamkhi Louai

Plateforme de gestion du secteur de la santé

Salhi Chaima , Bakour Abdel Ali

Development of a Web Application Dedicated to the Sale and Rental of Real Estate

Hemmak Ziad

Virtual Try-on E-Commerce

Saifi Abdelhamid

Fouille d’interactions dans les réseaux complexes

Benseddik Samir , Ghodbane Amir

Plateforme de vente en ligne de produits reconditionnés

Manel Mekdour , Chames Elassil Guettouche

PLATEFORME DE LOCATION DES OUTILS ENTRE PARTICULIER

HOUSSAM EDDINE BOUTCHICHA

EASY SHOT the Search for the Perfect Photographer

SACI FATIMA EZ ZAHRA , LAKHEDAR HAMINA SAADIA

Consultations médicales et orientations cliniques

Bourenane Nour El Houda , Felliachi Radjiya Somia

Learning Management System Platform and Web Application for a Private School

DEBBAH, RANIA , RAHMANI SAFA

APPLICATION WEB POUR LA GESTION D 'UNE AGENCE DE LOCATION DE VOITURES

Rezkallah Roqiya , Athamnia Dounia

Plateforme de Consultations Médicale pour l'accès aux soins de santé

Faysal TOUIL , Riadh LAMRAOUI

Mise en place d’un système d’information décisionnel à la société des ciments de Ain El Kebira

MESSARA Norhane , BENZIANE Hiba

Opportune : Social Networking Mobile Application

Zaoui Islam , Mohammed Lamine Bennia, Souleyman Karim Barkat Abdelmoumin

DEVELOPMENT OF AN E-LEARNING PLATFORM

Mihoubi Samir , Bouhafes,Nadji

Application (Desktop/mobile) pour Tram M'sila

SOUKHAL, Ramdane , DJADJA, Mohamed Amine

Conception et Mise en Ligne d’une Plateforme de Covoiturage en Algérie

Hamouda, Ibtissam , Naami, Cheyma

Une plateforme pour la conception publicitaire

TAHRI, Zakaria , FEIDJEL, Ismail

DESIGN & IMPLEMENTATION OF A COMPREHENSIVE FREELANCING AND SELF-EMPLOYING PLATFORM

Biraz, Zina , Zerd, Djamila

Plateforme pour anti gaspillage

BENABBAS Wassim

“Addressing the challenges of plant diseases diagnosis based on deep learning”

Chenane Ichraq

Analyse multidimensionnelle de documents

Rouag khouloud , Kara ilhem

Conception et Réalisation d’un Système d’Information Décisionnel pour l’université Mohamed Boudiaf M’SILA

Imad Eddine Dahmane , Belhadj Khaled

Détection de fausses Informations dans les réseaux Sociaux

Gasmi Lynda

Deep Lerning for Face Recognition

Boudraf Aissa

Développement d'une application mobile pour aider les randonneurs

Zahar Youcef

Mobile Application for medical Assistance

LOGOUI Khaoula Zineb

Réseau Social pour le suivi de patients diabétiques

Lograda Nassira

La détection de communautés dans les réseaux sociaux

Zouaoui Aya

Trust Evaluation and Prediction in Online Social Network

Hamouche Rabah

Système de Reconnaissance des personnes par la biométrie multimodale

BAABCHA Halima

Sécurisation des données biométriques.

Laifa Meriem

Trust and related concept in social networks

Loucif Hamza

L'analyse de l'influence dans les réseaux sociaux

This study investigates influence diffusion in online social networks (OSNs) through a comprehensive analysis of centrality measures and diffusion models using the Higgs Twitter dataset. We model OSNs as directed graphs, focusing on strongly connected components (SCCs) and weakly connected components (WCCs). Seven centrality measures (out-degree, in-degree, betweenness, closeness, eigenvector, PageRank, and Katz centrality) are calculated to identify key influential nodes. The top-ranked nodes are then subjected to influence diffusion simulations using three models: Linear Threshold (LT), Independent Cascade (IC), and Susceptible-Infected-Recovered (SIR) across three types of activity networks with different structural characteristics. Our findings reveal significant variations in centrality performance depending on network topology and diffusion dynamics. This methodology integrates structural network analysis with dynamic diffusion modeling to evaluate the effectiveness of influence spread. The experimental results show that out-degree and betweenness centralities are most effective for influence propagation, with the SIR model supporting sustained diffusion. The experimental results reveal that out-degree and betweenness centralities are the most effective measures for influence propagation, with out-degree being particularly impactful for initiating diffusion. The SIR model demonstrated superior efficacy for sustained influence spread, aligning more closely with real-world influence dynamics. Additionally, analyzing influence propagation within WCCs enables more computationally efficient identification of key influencers, without significant loss in accuracy. This work offers actionable insights for influence modeling and provides a practical methodology for selecting centrality measures tailored to specific diffusion scenarios. It explores the influence diffusion across different platforms, enabling researchers to assess and compare user impact by offering a detailed examination of network structures, key node significance and influence diffusion.

Citation

Samir Akhrouf , , (2025-12-18), Evaluating Centrality-Based Seed Node Strategies for Influence Diffusion in OSNs: A Study across SCC, WCC and Full Networks using SIR, LT and IC Diffusion Models, Informatica, Vol:49, Issue:22, pages:18, Slovenian Society Informatika

Spammers in online social networks continuously adapt their strategies, making detection a challenging and dynamic task. While traditional machine learning models and static deep learning approaches such as CNNs achieve good performance, they often fail to capture the temporal evolution of user behavior and network interactions. In this paper, we propose a novel deep learning framework for dynamic spammer detection that combines Principal Component Analysis (PCA) for feature reduction, Convolutional Neural Networks (CNNs) for local content feature extraction, and Temporal Graph Embeddings (TGEs) to capture evolving interaction patterns over time. Unlike prior static models, our approach explicitly models the dynamics of user behavior and relational changes in the social graph. Experiments conducted on benchmark Twitter datasets demonstrate that our hybrid PCA CNN–TGE model significantly outperforms classical baselines (ANN, CNN, SVM) and static hybrid models, achieving an F1-score of 94 %. The results highlight the importance of temporal graph learning for robust and adaptive spammer detection in social networks.

Citation

Hemza Loucif , Samir Akhrouf , ,(2025-11-25), Dynamic spammer detection using deep learning with temporal graph embeddings,Second International Workshop on Machine Learning and Deep Learning,Mohamed Boudif University M'Sila, Faculté MI

Early detection of Diabetic Retinopathy (DR) is essential to reduce the risk of vision loss. This study introduces a novel framework for DR detection using a Particle Swarm Optimized Autonomous Learning Multiple Model (PSO-ALMMo) system. The proposed approach integrates the adaptive learning capability of the ALMMo* system with particle swarm optimization (PSO) to enhance classification accuracy and model interpretability. The proposed method uses PSO to optimize both antecedent and consequent parameters of the ALMMo* model, enabling high performance while maintaining the ability to learn incrementally from new data without retraining. Hybrid feature extraction techniques are applied to retinal fundus images before classification. Experiments were conducted on the newly introduced LISIA dataset as well as three benchmark datasets: Messidor-2, APTOS 2019, and IDRID. The PSO-ALMMo* system achieved 98.2% accuracy on Messidor-2, 99.7% on APTOS 2019, and 99% on IDRID. On the LISIA dataset, it maintained consistent performance across all DR severity levels. The model facilitates real-time learning and generates interpretable outcomes, owing to its prototype-based structure. These results indicate that the proposed system is well-suited for clinical environments to support early and accurate DR screening.

Citation

Mohamed Chatra , Samir Akhrouf , abdelouahab.attia@univ-bba.dz, PARTICLE SWARM OPTIMIZED ALMMO* FOR INTERPRETABLE AND ACCURATE DIABETIC RETINOPATHY DETECTION, zineb.maaref@univ-bba.dz, , (2025-10-30), PARTICLE SWARM OPTIMIZED ALMMO* FOR INTERPRETABLE AND ACCURATE DIABETIC RETINOPATHY DETECTION, JZU NATURAL SCIENCE, Vol:56, Issue:10, pages:23, Journal of Zhengzhou University-Natural Science

Traffic management, urban planning, and emergency management cannot be efficiently done without crowd simulation. This paper proposes a Behavioral Clustering Method (BCM), which tackles the problem of forming crowds in clusters or subgroups based on fundamental behaviors so that congestion is minimized during effective evacuation processes. We designed BCM based on synthetic data obtained from the simulation of the evacuation of a crowd in high-risk situations. Our method regards pedestrians as intelligent agents and predicts key behavioral aspects of future crowd evacuations before they occur. We use cluster analysis on those movement and behavioral data for building as well as evacuation-friendly control strategies by clustering people into subgroups of behavioral similarity. The credibility of the model is validated through Python-based animations to detect and rectify errors. Results from simulation performance evaluations indicate that BCM is successful in modeling the evolution of crowd behavior at the time of evacuation.

Citation

Mohamed Chatra , ghenabzia ahmed , Samir Akhrouf , Mustapha Bourahla , , (2025-08-31), A Simulation-Based Behavioral Clustering Method for Crowd Dynamics Evacuation Analysis, HAUT, Vol:23, Issue:8, pages:20, Open Access

Influence maximization in large-scale social networks faces challenges of scalability and community-aware diffusion. As Online social networks (OSNs) are deeply integrated into our daily lives, particularly with the continuous expansion of web services and mobile technologies. The information shared by individuals through social connections directly impacts our beliefs and behaviors. Consequently, identifying influential nodes in complex OSN structure has attracted considerable attention in a wide range of applications such as viral marketing and managing rumors. However, traditional centrality-based methods often fail to achieve optimal influence spread in large-scale, real-world networks due to their high computational complexity and limited consideration of community structures. To address these limitations, in this paper, we propose a novel hybrid approach that combines community detection with influence score calculation. The method evaluates influence within communities (intra-community) using centrality measures, and across communities (inter-community) by identifying bridging nodes, producing a total influence score for each node. We implement the approach using the Louvain algorithm for community detection and a hybrid bridging score to capture inter-community influence. The proposed method is evaluated using the Susceptible-Infected-Recovered (SIR) and Independent Cascade (IC) diffusion models on three real-world datasets. Experimental results demonstrate that our method identifies influential nodes capable of propagating influence more rapidly and effectively compared to existing techniques, demonstrating its scalability and effectiveness for large-scale networks.

Citation

Samir Akhrouf , halima.baabcha@univ-bba.dz, meriem.laifa@univ-bba.dz, , (2025-07-31), Hybrid Community-Driven Influence Maximization in Large Social Networks, Ingénierie des Systèmes d'Information, Vol:30, Issue:7, pages:13, International Information and Engineering Technology Association

Federated learning (FL) allows distributed devices to jointly train a global model while safeguarding the privacy of their local data. However, selecting and securing clients, especially in environments with potentially malicious participants, remains a critical challenge. This study proposes an innovative participant selection method to enhance both security and efficiency in centralized and decentralized FL frameworks. In the centralized framework, this method effectively excludes clients with weak privacy protections and optimization capabilities, thus increasing overall system security. For decentralized FL, a blockchain-supported approach is introduced, which further strengthens the robustness of the system. Using a dynamic role assignment algorithm, roles such as worker, validator, and miner are allocated based on security and performance metrics for each training round. The findings show that this method performs on a par with the scenarios free of malicious clients, demonstrating the value of blockchain technology in improving FL protocols. By addressing security vulnerabilities and improving training efficiency, this research contributes to the development of more secure and efficient FL systems, underscoring the importance of advanced participant selection and role assignment strategies.

Citation

WAFA Bouras , Mohamed Benouis , BRAHIM Bouderah , Samir Akhrouf , kameleddine.heraguemi@ensia.edu.dz, , (2025-06-16), Optimizing Security and Performance in Blockchain-Enhanced Federated Learning Through Participant Selection with Role Determination, Computing and Informatics, Vol:44, Issue:3, pages:35, OJS/PKP

AI has transformed the research process, enhancing efficiency and accuracy in tasks like literature reviews, data analysis, and writing. We will explore how AI tools streamline research workflows, highlight best practices for integrating AI into academic work. Let's discover how AI can transform your research journey.

Citation

Samir Akhrouf , ,(2025-04-16), Artificial Intelligence Tools in Academic Research,Pedagogical Days (Ped Days 2025),Mohamed Boudif University M'Sila, Faculté MI

This paper proposes a novel approach to accelerate association rule mining using the Reptile Search Algorithm (RSA) in conjunction with GPU-based parallel processing. Traditional association rule mining techniques can be computationally expensive, especially with large datasets. By utilizing the inherent parallelism of Graphics Processing Units (GPUs), we significantly speed up the fitness evaluation process, a core component of the Reptile Search Algorithm. Our results show a marked improvement in the performance of RSA on large datasets, making it feasible for real-time or large scale applications such as market basket analysis, healthcare for drug interaction analysis, and web usage mining. We also analyze the impact of various GPU optimizations and present a comparison with CPU-based execution.

Citation

Abderrahim Boukhalat , Mohamed Benouis , BRAHIM Bouderah , Samir Akhrouf , kameleddine.heraguemi@univ-msila.dz, ,(2024-12-10), Parallel Association Rules Mining Using GPUs and Reptile Search Algorithm,The Sixth International Symposium on Informatics and Its Applications (ISIA),Université Mohamed Boudiaf M'Sila

Recently, several studies have used deep convolutional neural networks (DCNN) for plant disease classification based on leaf symptoms images. Most of these studies use the PlantVillage dataset, which contains simple images with simple backgrounds. This has led to classifiers achieving high accuracy in academic settings but underperforming in practical scenarios where images taken by farmers are much more complex. In this paper, we propose a data augmentation (DA) method that transforms simple images from the PlantVillage dataset into more complex images, thereby bridging the gap between academic results and practical performance. Our technique focuses on making the images more realistic by generating images containing multiple leaves and complex backgrounds, such as soil. To evaluate the impact of our proposed method, we trained DCNN models using various augmentation methods and tested them on a dataset containing highly complex images comparable to those taken by farmers, posing a substantial challenge. The experimental results showed that models trained on images generated by our method outperformed traditional data augmentation methods, such as geometric transformations. Moreover, our method is competitive with Generative Adversarial Networks (GANs) without requiring any training phase.

Citation

Samir Akhrouf , wassim.benabbas@univ-bba.dz, mohamed.brahimi@ensia.edu.dz, bilal.fortas@univ-bba.dz, , (2024-10-01), Improving plant disease classification using realistic data augmentation, Multimedia Tools and Applications, Vol:83, Issue:38, pages:19, Springer Nature

Early detection of Diabetic Retinopathy (DR) is crucial for effective intervention, particularly given its often asymptomatic nature in initial stages. Automated detection systems offer significant advantages, such as improving screening efficiency, extending healthcare accessibility to remote regions, and facilitating proactive disease management. Introducing an innovative framework for retinal image analysis that combines explainable deep learning, hybrid feature extraction, and advanced data augmentation to optimize performance and interpretability for clinical applications. The xDNN model, trained on the MESSIDOR-2 dataset, achieved an average recall of 98%, precision of 98.2%, F1-score of 98%, and accuracy of 98.2%. When extensively trained on the APTOS 2019 dataset, the model delivered outstanding results with an average precision, recall, and F1-score of 99%, and an accuracy of 99.7%. Additionally, the model’s performance on the IDRID dataset was remarkable, with average precision, recall, F1-score, and accuracy all reaching 99%. Noteworthy is our method impressive average AreaUnder the Curve (AUC) of 99.8%, affirming its consistent and exceptional performance across all classes of diabetic retinopathy. This underscores the xDNN Classifier’s potential as a valuable tool for precise and reliable diabetic retinopathy detection. It holds substantial promise for elevating clinical diagnosis and enhancing treatment outcomes within the field of ophthalmology.

Citation

Samir Akhrouf , oualid.mecili@univ-bba.dz, hadj.barkat@gmail.com, attia.abdelouahab@gmail.com, farid.nouioua@gmail.com, , (2024-08-23), An efficient explainable deep neural network classifier for diabetic retinopathy detection, International Journal of Computers and Applications, Vol:46, Issue:9, pages:17, Taylor and Francis

Association rule mining (ARM) is a very popular, engaging, and active research area in data mining. It seeks to find valuable connections between different attributes in a defined dataset. ARM, which describes it as an NP-complete problem, creates a fertile field for optimization applications. The Reptile Search Algorithm (RSA) is an innovative evolutionary algorithm. It yanks stimulation from the encircling and hunting conducts of crocodiles. It is a well-known optimization technique for solving NP-complete issues. Since its introduction by Abualigah et al. in 2022, the approach has attracted considerable attention from researchers and has extensively been used to address diverse optimization issues in several disciplines. This is due to its satisfactory execution speed, efficient convergence rate, and superior effectiveness compared to other widely recognized optimization methods. This paper suggests a new version of the reptile search algorithm for resolving the association rules mining challenge. Our proposal inherits the trade-off between local and global search optimization issues demonstrated by the Reptile search algorithm. To illustrate the power of our proposal, a sequence of experiments is taken out on a varied, well-known, employing multiple comparison criteria. The results show an evident dominance of the proposed approach in the front of the famous association rules mining algorithms as well as Bees Swarm Optimization (BSO), Bat Algorithm (BA), Whale Optimization Algorithm (WOA), and others regarding CPU time, fitness criteria, and the quality of generated rules.

Citation

Abderrahim Boukhalat , Mohamed Benouis , BRAHIM Bouderah , Samir Akhrouf , kameleddine.heraguemi@univ-msila.dz, , (2024-06-01), Reptile Search Algorithm for Association Rule Mining, International Journal of Computing and Digital Systems, Vol:14, Issue:1, pages:16, University of Bahrain

To the best of current knowledge, the performance of federated learning predominantly depends on the efficiency of the aggregation server scheme utilized to consolidate model parameters received from distributed local devices. However, in practical scenarios, the global server often faces single-point failures due to four major issues: 1) variations in data distribution settings, such as independent identical distribution (IID) or non-independent identical distribution; 2) communication overhead; 3) limitations in hardware and resource storage availability; and 4) diverse participant participation behaviors. To address the latter concern, limited research has endeavored to establish a correlation between these heterogeneous settings and federated learning performance by analyzing different aspects of participant behavior. Inspired by the absence of a definitive verdict regarding the relationship between the global server and participant behavior, this paper investigates the aspect of participant selection methods and conducts a detailed comparative study among various participant selection methods

Citation

WAFA Bouras , Mohamed Benouis , Samir Akhrouf , brahim.bouderah@univ-msila.dz, ,(2024-05-13), Analysis Study of Participant Selection Methods in Federated Learning,ICEEAC’2024,Setif university

في عالمنا المتسارع اليوم، أصبحت منصات التواصل الاجتماعي وسيلة أساسية للتواصل والتفاعل بين الناس. فهي تتيح لنا الاتصال بأصدقائنا وعائلتنا، والتعرف على ثقافات جديدة، والمشاركة في المناقشات والأفكار المختلفة.

Citation

Samir Akhrouf , ,(2024-05-08), مخاطر استعمال منصات التواصل الاجتماعي,اليوم التحسيسي ضد الأخطار المتعلقة باستعمال وسائل التواصل الاجتماعي الأربعاء 08 ماي 2024,Université Mohamed Boudiaf M'Sila

Diabetes mellitus, a widely recognized chronic disease often termed a silent killer, disrupts insulin production, leading to elevated blood sugar levels and complications affecting organs including the eyes, kidneys, and nerves. Despite significant research interest, diabetes classification-based machine learning methods suffer from low performance due to missing data and imbalanced class data. Thus, this paper introduces a new method for Diabetes Mellitus classification for Imbalanced data with Missing values (DMIM). The proposed approach involves these key steps. Firstly, missing values are preprocessed using the Isolation Forest method that provides data normalization. Then, an adaptive synthetic sampling technique, Smote T-link, is employed to counter the impact of class imbalance. Finally, we utilized a Random Forest (RF) classifier and conducted experiments on the Pima Indians diabetes dataset from the University of California. Our proposed approach yielded highly accurate predictions, as evidenced by impressive performance metrics such as an Area Under the Curve (AUC) of 97.2%, Specificity (SP) of 91.7%, and Recall (Rec) of 92.3%, Further comparison between other ML methods including k-nearest Neighbors, Decision Trees, Random Forest, AdaBoost, Naive Bayes, XGBoost, Extra Trees. The results demonstrate the effectiveness and superiority of our proposed method and improved diabetes classification performance compared to existing techniques.

Citation

Samir Akhrouf , mecilioualid@gmail.com, hadj.barkat@gmail.com, farid.nouioua@gmail.com, attia.abdelouahab@gmail.com, ,(2024-04-24), Efficient Machine Learning Approach for Diabetes Mellitus Disease Prediction,Conference: 2024 6th International Conference on Pattern Analysis and Intelligent Systems (PAIS),El OUED Algérie

The emergence of so many online social networks (OSNs), such as Twitter, YouTube...etc, has significantly altered the global landscape of human connections and interactions. OSNs have become an integral part of people’s lives and have led to an explosion in the volume and variety of information available for research and analysis. In social network analysis (SNA), Influential user identification in OSN involves detecting users who have a significant impact on information dissemination and network dynamics is a fundamental research task. It plays an important role in different domains such as political movements, health applications, innovation dissemination, news spreading, and viral marketing. In this paper, we introduce a theoretical framework for identifying influential users based on both structural and behavioral factors of online social networks. We first present the problem statement and provide a review of pertinent background knowledge. Subsequently, our proposed framework is explained and discussed.

Citation

Samir Akhrouf , Baabcha Halima, Meriem Laifa, ,(2023-12-06), A Conceptual Framework for Identifying Influential Users in Online Social Networks for Viral Marketing,THE 6TH INTERNATIONAL HYBRID CONFERENCE ON INFORMATICS AND APPLIED MATHEMATICS IAM’23,Guelma Algeria

Online social networks (OSNs) have become an integral part of people's daily lives, resulting in a vast amount and diversity of information available for research and analysis. Detecting influential users is a significant research topic in social network analysis (SNA), and it‘s played a crucial role in information dissemination within OSNs. It has various applications, including viral marketing. In this short paper, we describe our thesis subject by explaining the context, research objectives, and motivations. Subsequently, we present our proposed approach and research methodology.

Citation

Samir Akhrouf , Baabcha Halima, Laifa Meriem, ,(2023-10-03), Influencers Detection in Online Social Network,The 1st National Conference on New Educational Technologies and Informatics NCNETI'23,Guelma Algeria

Frequently item-sets mining, also known as FIM, is a data mining technique used to extract useful knowledge from datasets. FIM is plagued by a number of issues, including high storage charges, a huge time and memory consumption. Classical FIM algorithms suffer with the great number of generated itemsets which contains useless items. And presupposes that all itemsets have the same importance. To deal with these limitations, high-utility itemsets (HUI) is proposed, which are as subset of FIM with the consideration of Utility or profit measure. In the last decade, evolutionary approaches become a trend to solve HUIs mining problem (EA). This paper explores the use of evolutionary techniques in HUIs mining. Moreover, we present an evolutionary techniques-based HUIs mining classification including single objective, multi-objective and Hybrid optimization techniques. This article provides a comparative examination of methodologies and discusses theoretical features of a wide variety of algorithms that are inspired by nature.KeywordsHUIs miningBio-inspired algorithmsEvolutionary approaches

Citation

Samir Akhrouf , Abderrahim Boukhalat, Kameleddine Heraguemi, Mouhamed Benouis, Bouderah brahim, ,(2023-07-15), A Survey on Using Evolutionary Approaches-Based High-Utility Itemsets Mining,Artificial Intelligence Doctoral Symposium,Alger

Social networks such as Facebook, Twitter, etc. have dramatically increased recently. These databases are massive and their use is time-consuming. In this work, we present an optimal calculation in graph mining for link prediction to reduce the runtime. For that purpose, we propose a novel approach that operates on the connected components of a network instead of the whole network. Thanks to this decomposition, we show that the results of all link prediction algorithms using local and path-based similarity measures can be achieved with much fewer computations and hence within a much shorter runtime. We show that this gain depends on the distribution of nodes in components and may be captured by the Gini and the variance measures. We propose a parallel architecture of the link prediction process based on the connected component's decomposition. To validate this architecture, we have carried out an experimental study on a wide range of well-known datasets. The obtained results clearly confirm the efficiency of exploiting the decomposition of the network into connected components in link prediction.

Citation

Samir Akhrouf , , (2023), Fast approach for link prediction in complex networks based on graph decomposition, Evolving Systems, Vol:1, Issue:1, pages:18, Springer Nature

Given the still-existing restrictions of COVID-19, blended learning is undoubtedly becoming a better-fitting strategy for higher education institutions in underprivileged countries. Acknowledging the current changes in higher education, this study aims to investigate the elements that influence students’ satisfaction and their future preferences regarding blended learning in Algeria. A total of 782 questionnaires were collected from different Algerian universities. A structural Equation Modeling (SEM) analysis was conducted to investigate the relationship among the latent variables of the proposed theoretical model. Moreover, an unsupervised sentiment analysis approach was applied to analyze the qualitative data received as feedback from the participants. The results show that students perceived ease of use and perceived usefulness of blended learning had a significant positive impact on their satisfaction. Similarly, satisfaction had a positive influence on students’ future preferences regarding blended learning. In turn, students’ perceived ease of use and usefulness indirectly affected their future preferences, mediated by satisfaction. Additionally, qualitative data echoed students’ eagerness to adopt more advanced learning technologies and what obstacles currently stand in their way. The contribution of this study is to reflect the current situation of blended learning adoption in developing countries and to support future curriculum planning and development. It can also help teachers, students, and policymakers to make better decisions and recommendations for an improved and more sustainable learning and teaching environment in the future.

Citation

Samir Akhrouf , Meriem Laifa, Roya Imani Giglou, , (2023), Blended Learning in Algeria: Assessing Students’ Satisfaction and Future Preferences Using SEM and Sentiment Analysis, Innovative Higher Education, Vol:11, Issue:2, pages:27, Springer Nature

The conversation on the effectiveness of blended learning experience in Algerian higher education has not received the necessary attention compared to developed countries. Moreover, there is a scarcity of research that discusses the impact and benefits of the pandemic-induced changes in Algerian higher education. The primary goal of this paper is to initiate a deeper conversation about blended learning in Algerian Higher Education using an exploratory research method based on empirical evidence and quantitative data. It examines teachers’ and students’ satisfaction with blended teaching and learning at Algerian universities since the start of Covid-19. There was a total of 782 students and 82 teachers involved with valid questionnaire responses, from two different Algerian public universities. The data were examined using descriptive and inferential statistical analysis, such as independent t-test, Mann-Whitney U-test, Kruskal-Wallis test, and one-way ANOVA. The results indicate low satisfaction with blended learning on both sides: students and teachers. Furthermore, the data do not suggest differences in satisfaction between teachers and students, either in terms of the program taught or the university to which they belong. However, there were significant differences among students based on their age and their level of study. The paper also discusses the values, benefits, and restrictions of the mandatory blended learning approach applied by the Algerian ministry since March 2020. Implications and recommendations for both practitioners and researchers are additionally addressed in this study to make informed pedagogical choices in the future.

Citation

Samir Akhrouf , Meriem Laifa, Mohdeb Djamila, Belazzoug Mouhoub, , (2023), SATISFACTION OF STUDENTS AND TEACHERS WITH BLENDED LEARNING DURING COVID-19 PANDEMIC: AN ALGERIAN CASE STUDY, International Journal of Learning Technology, Vol:11, Issue:2, pages:24, Inderscience

These days, face recognition systems are widely being employed in various daily applications such as smartphone unlocking, tracking school attendance, secure online bank transactions, smarter border control, to name a few. In spite of the remarkable progress, face recognition systems still suffer from occlusions, light variations, camera types, and their resolutions. Face recognition is still a dynamic research field. In this paper, we propose an efficient face recognition system based on the Gabor filter bank and a deep learning method known as Sparse AutoEncoder (SAE). The main aim of the proposed system is to improve the features extracted by the Gabor filter bank using the SAE method. Then, these enhanced features are subjected to features reduction using principal component analysis and linear discriminant analysis (PCA+LDA) technique. Finally, the matching stage is accomplished via cosine Mahalanobis distance. Experiments on seven publicly available databases (i.e., JAFFE, AT&T, Yale, Georgia Tech, CASIA, Extended Yale, Essex) show that the proposed system can achieve promising results with the combination of Gabor and SAE, as well as outperform previously proposed methods.

Citation

Samir Akhrouf , Hammouche Rabah, Attia Abdelouahab, Akhtar Zahid, , (2022), Gabor Filter Bank with Deep Autoencoder based Face Recognition System, Expert Systems with Applications, Vol:197, Issue:1, pages:10, ScienceDirect

This paper present a novel system for person authentication based on score level fusion of Minor and Major dorsal finger knuckle patterns. In the proposed method, the adaptive single scale retinex method is used to extract the reflectance and the illumination of Major and Minor traits respectively, also the binarized statistical image features method is used to extract normalized histogram features. Furthermore, the Cosine Mahalanobis distance is used in the matching stage. Moreover, a multi-biometric system-based score level fusion has been proposed. In an attempt to enhance performance recognition, the symmetric sum-based rules based on triangular norms are applied. The system is evaluated on the publically Minor/Major knuckle database. Experiments conducted on this database achieved good results. Besides, the proposed system outperforms the previous methods given in the state of the art.

Citation

Samir Akhrouf , Hammouche Rabah, Attia Abdelouahab, , (2022), Score level fusion of major and minor finger knuckle patterns based symmetric sum-based rules for person authentication, Evolving Systems, Vol:13, Issue:2, pages:15, Springer Nature

Palmprint recognition systems have been extensively studied over the past two decades because of their unique, accurate, and stable biometric features. Many researchers have investigated the two-dimensional (2D) palmprint recognition that contains texture information, but the 2D palmprint image does not contain three-dimensional (3D) depth information. To beat the limitations associated with a 2D palmprint recognition system; this paper proposes using both 2D and 3D palmprint features for a personal recognition system. The BSIF and GIST descriptors are utilized for feature extraction from the 2D and 3D palmprint. Then, the PCA + LDA technique is used to reduce the dimensionality of feature vectors. Next, the matching process is done using the Cosine distance. Finally, a score-level fusion was applied to get a final matching score using a fuzzy connective method based on the linear combination of triangular norms (T-norms and T-conforms). Also, the proposed method is compared to several techniques of score fusion, including Sum, Min, Max, and Symmetric Sum based on triangular norms. The system was evaluated on the publicly available Hong Kong-PolyU 2D + 3D palmprint database, and the results show the proposed system's efficiency.

Citation

Samir Akhrouf , Attia Abdelouahab, Rabah Hammouche, Zahid Akhthar, , (2022), A fuzzy connective score fusion technique for 2D and 3D palmprint authentication system, Evolving Systems, Vol:13, Issue:4, pages:12, Springer Nature

This paper presents a new version of one of the models that we have proposed to measure the influence of web users in social networks like Facebook. The new version enhanced the previous one through the incorporation of two substantial modules, namely the global impression that the postings of the potential influencer get from his followers and the entropy which manifests the quantity of information carried by those postings. The comparison of our model with the precedent version and the PageRank benchmark has shown the effectiveness of our updates and the importance of incorporating the entropy and the global impression factors in its formulation.

Citation

Hemza Loucif , Samir Akhrouf , ,(2021), A Recursive Model to Measure Influence in Subscription Networks: A Case Study using Twitter.,International Conference on “Managing Business through Web Analytics" ICMBWA2020,,Université Djilali Bounaama – Khemis Miliana,

The advent of online social networks has been one of the most exciting events in the last years. Their richness in content can be leveraged for analysis; the richness of this network provides unprecedented opportunities for data analytics. Social influence analysis is one of the major research topics in social networks analysis. It can be used in many applications including viral marketing. Modeling influences diffusion and influence maximization in social networks is an important challenging in the area of social network analysis. Furthermore the identification of influential nodes in a social network. There is an abundance of research that has been done regarding influence in online social networks for various applications, especially in a viral marketing context. In this paper, a survey of the most and recent important methods for influence modeling, maximization, and influential nodes detection. We aim to guide novice researchers by providing them with the leading works on social influence analysis in the context of viral marketing.

Citation

Samir Akhrouf , Baabcha Halima, Laifa Meriem, ,(2021), Social Influence Analysis in Online Social Networks for Viral Marketing: A Survey,International Conference on “Managing Business through Web Analytics" ICMBWA2020,,Université Djilali Bounaama – Khemis Miliana,

During the last decade, traditional interpersonal relationships have been replaced by online communications due to the rapid evolution of online social networks. Trust plays an important role in enhancing interactions and ensuring security in such online communications. The decision to trust a user or not depends on a number of factors such as the similarity between users. The concept of user similarity is used in various fields for different aims. This paper presents an overview of the user similarity concept in an online trust-related context. We present various similarity metrics used by researchers in this field, and we provide a literature classification, including characteristics, data, and applications based on existing well-established studies. Finally, we review related work that focuses on both similarity and trust in online social networks.

Citation

Samir Akhrouf , Zouaoui Aya, Laifa Meriem, ,(2021), User Similarity and Trust in Online Social Networks: An Overview,International Conference on “Managing Business through Web Analytics" ICMBWA2020,,Université Djilali Bounaama – Khemis Miliana,

The authentication of individuals based on Finger Knuckle print (FKP) is a very interesting system in the biometric community. In this paper, we introduce a biometric authentication system based on the FKP trait which consists of four stages. The first one is the extraction of the Region of Interest (ROI). The Phase Congruency method with Gabor filters bank descriptors has been used in the feature extraction stage. Then to enhance the performance of the proposed scheme the Principle Component Analysis (PCA) + Linear Discriminant Analysis (LDA) method has been used in the dimensionality reduction stage. Finally, cosine Mahalanobis distance has been used in the matching stage. Experiments were conducted on the FKP PolyU Database which are publicly available. The reported results with comparison to previous methods prove the effectiveness of the proposed scheme, as well as the given system can achieve very high performance in both the identification and verification modes.

Citation

Samir Akhrouf , Hammouche Rabah, Attia Abdelouahab, , (2020), A NOVEL SYSTEM BASED ON PHASE CONGRUENCY AND GABOR - FILTER BANK FOR FINGER KNUCKLE PATTERN AUTHENTICATION, ICTACT JOURNAL ON IMAGE AND VIDEO PROCESSING, Vol:10, Issue:3, pages:1-7, JOURNAL ON IMAGE AND VIDEO PROCESSING

Diabetic Retinopathy (DR) is one of the mainly causes of visual loss worldwide. In fact, DR is leading source of impaired vision in people between 25 and 74 years old. DR exists in wide ranged and its detection is a challenging problem. The gradual deterioration of retina leads to DR with several types of lesions, including hemorrhages, exudates, micro aneurysms, etc. Early detection and diagnosis can prevent and save the vision of diabetic patients or at least the progression of DR can be slowed down. The manual diagnosis and analysis of fundus images to substantiate morphological changes in micro aneurysms, exudates, blood vessels, hemorrhages, and macula are usually time-consuming and monotonous task. It can be made easy and fast with the help of computer-aided system based on advanced machine learning techniques that can greatly help doctors and medical practitioners. Thus, the main focus of this paper is to provide a summary of the numerous methods designed for discovering hemorrhages, microaneurysms and exudates are discussed for eventual recognition of non-proliferative diabetic retinopathy. This survey will help the budding researchers, scientists, and practitioners in the field.

Citation

Samir Akhrouf , Abdelouahab Attia, Zahid Akhtar, Sofiane Maza, , (2020), A SURVEY ON MACHINE AND DEEP LEARNING FOR DETECTION OF DIABETIC RETINOPATHY, JOURNAL ON IMAGE AND VIDEO PROCESSING, Vol:11, Issue:2, pages:8, ICTACT

Most minutiae-based matching techniques use local static descriptor to emphasize their discrimination power between ambiguous minutiae features. Despite the improvements shown by these methods, still some challenging problems limit their performance in particular the presence of spurious (added) minutiae and the missing of genuine minutiae. The added/missed minutiae problem misleads the matching process and complicates the correspondence decision. A general countermeasure of most existing matching algorithms is to add a global matching consolidation step that results in augmented time complexity. In this work, a dynamic minutiae-based descriptor is described. Its particularity resides in its ability to adjust a minutia features according to the correspondent minutia context in the second print. The minutiae of the input fingerprint are arranged to get a global stable geometric structure called minutiae-based polysegment structure (MPS). It permits to detect added/missed minutia once compared to another MPS structure. Whenever the matching algorithm detects an added minutia in an MPS, it inserts a virtual minutia in the second to update the minutia descriptor context and, hence, let propagating the similarity. This increases the chance to maximize the number of genuine paired minutiae. Furthermore, the MPS structure reduces enormously the minutiae space to be tested; the matching algorithm complexity is near O(n log(n)). Experiments of the proposed algorithm are conducted on the public database FVC2002.

Citation

Samir Akhrouf , Belhadj Foudil, ,(2019), A fast minutiae-based fingerprint-matching algorithm using a dynamic descriptor and virtual minutiae insertion,ICPAR 2019 International Conference on Pattern Analysis and Recognition,Tebessa

Medical care has always presented quite wide ranged and challenging problems. However, machine learning techniques and methods as well as deep learning never stopped evolving and tackling those challenges issued by medicine, medical and health care. In order to have a more close up look on how machine learning and deep learning has been affecting medical care in general, we review in this paper some machine learning and deep learning techniques used in a variety of medical care sections such as medical imaging, medical decision, diagnostic, medical records and big data, and disease prediction.

Citation

Samir Akhrouf , Chalabi Nour El Houda, Attia Abdelouaheb, ,(2019), Machine learning & Deep learning: Recent Overview in Medical Care,IMCL 2019,Thessaloniki Greece

Diabetes is a group of metabolic diseases in which a person has high blood sugar, either because the body does not produce enough insulin, or because cells do not respond to the insulin that is produced. The constant hyperglycemia of diabetes is related to long-haul harm, brokenness, and failure of various organs, particularly the eyes, kidneys, nerves, heart, and veins. This last can be monitoring by doctors over the internet and especially social net- works.In this context, having a healthcare social media strategy is no longer optional, it’s a requirement. With the right strategy, social media will become a powerful tool to build trust, reach more patients, and spread valuable medical information. Furthermore, advances in mobile technology and the widespread use of smartphones and tablets will make an improvement in healthcare services at a rapid pace.In this paper, we propose a system that supports a community of diabetes people, their families and friends, doctors, nutritionist and anyone who might need or offer help and support to the community. Our objective is to develop a system that can predict, diagnose and monitor the diabetes or support diabetes self- management using machine learning algorithms and Big Data.

Citation

Samir Akhrouf , Mecili Oualid, Barkat Hadj, Nouioua Farid, Malek Rachid, ,(2019), The use of Big Data and Data Analytics in the prevention, the diagnosis and the monitoring of long-term diseases,IMCL 2019,Thessaloniki Greece

دور المنح الدراسية في تحسين العملية التعليمية الجامعية

Citation

Samir Akhrouf , Benfradj Zouina, ,(2019), دور المنح الدراسية في تحسين العملية التعليمية الجامعية,المؤتمر العربي الدولي التاسع لضمان جودة التعليم العالي,Beyrouth Liban

In the last decade advances in wireless technology led to the emergence of a wide range of applications in healthcare, mobile health care, social networking, etc. Among these applications, health care, mobile health and smart health are considered the most promising and most important, this is mainly due to their positive impact on the quality of life of patients. Furthermore, mobile health care technology helps improve the quality of care administered to patients providing clinical and critical care in ways that help them follow-up their disease. Therefore, patients can be treated and cared for from anywhere, at any time. In addition mobility helps patients track their recovery with remote monitoring tools and follow what doctors prescribed. Thus we can conclude that this area of research is very interesting and this is why we intend to propose this special session in order to gather researchers in the field of mobile health care, give them the opportunity to share their ideas and conceptual approaches and finally discuss the recent advances in this field.

Citation

Samir Akhrouf , ,(2019), Special Session Call for Papers Social Networks and Mobile Applications for Health care (SNMAH),IMCL 2019,Thessaloniki Greece

Social media users can easily be offended or hurt on those platforms, which leads to discomfort and health issues such as stress and anxiety. Forgiveness plays an important role to maintain healthy online relationships, which is the central constituent of social dynamics, from cooperation to social cohesion. While most prior studies have focused on analyzing forgiveness factors in offline settings using statistical methods, this study offers a new perspective using a two-staged approach whereby a research model was tested using structural equation modeling (SEM), and then the results were used as inputs for artificial neural network (ANN) and fuzzy logic (FL) models. An agent-based simulation was then performed to shed light on a possible use of the implemented models. Combining ANN and FL provided more accurate prediction results. In addition, simulation experiments call attention to the potential benefits of forgiveness in maintaining connectedness in a social network. The main purpose of this investigation was to evaluate the applicability of soft computing techniques on forgiveness prediction. Instead of relying on data mining techniques, we looked into questions that can improve our understanding of how society works in a digital age. In addition, this study provides an interesting example of a different and insightful way of doing computational social science that is useful to both researchers and practitioners.

Citation

Samir Akhrouf , Meriem Laifa, Ramdane Maamri, , (2018), Forgiveness and trust dynamics on social networks, Adaptive Behavior, Vol:26, Issue:2, pages:85-83, SAGE Journals

This article describes how interpersonal relationships structures and standards are evolving. By focusing on a social network context, this study examined different factors that can affect forgiveness decision of a victim of an online offense. In addition, it inspected whether the decrease of trust after an online-related offense can be affected by forgiveness. 323 participants took part in this study by completing a questionnaire that recorded different measurements. Structural equation modeling was used as the main technique for data analysis, and AMOS was used as a tool. Surprisingly, while empathy and commitment had no significant direct effect, results showed that the severity of the offense, its frequency and pretransgression trust are the main factors that influence forgiveness. Moreover, a victim's trust towards the transgressor decreased much more in the absence of forgiveness than in its presence. A valuable contribution of this article is in the prospect for related future research as well as the potential for applications that explore new techniques to facilitate forgiveness in the digital age.

Citation

Samir Akhrouf , Laifa Meriem, Roya Imani Giglou, Maameri Ramdane, , (2018), Forgiveness Predictors and Trust in a Digital Age, IJTHI, Vol:14, Issue:4, pages:23-42, IGI Global

Obesity can be classified as one of the most dangerous diseases in the world, as the number of obese people keeps on increasing. Obesity is related to many chronic diseases such as cancer, type 2 diabetes and heart disease, also to many factors such as eating habits and inactivity. This last can be caused by people’s addiction to Internet and especially social networks. This research introduces a new approach in the daily fight against obesity, by proposing a system that supports a community of overweight people, obese people, their families and friends, physicians, and anyone who might need help and support from this community or who can offer help to members of the community. In addition to the new system proposed a specific purpose social network to support this community was developed in order to help them discuss their problems, express their specific needs, share their experiences and motivate other members by sharing their experience with obesity… etc.

Citation

SamirAkhrouf , Benathmane Abdelkrim, Bouchachi Ramzi, Belayadi Yahia, ,(2018); Social Networking for Obesity Prevention and Eradication for Algerian teenagers,,Springer, Cham

As online communications are taking over traditional ones, new perspectives are needed to understand the effect of this shift on people’s interactions and behavior. In this short paper, we focus on the aspect of repairing broken relationships after a transgression or a conflict that occurs in an online environment, where we specifically highlight the importance of studying forgiveness in the context in which the conflict occurs. We also provide a brief discussion of possible paths for such research.

Citation

Samir Akhrouf , Laifa Meriem, Belayadi Yahia, Boubetra Djamel, ,(2018), Forgiveness and repairing the broken trust in digital age,IAJC 2018,Orlando USA

This paper aims to describe a new simplistic model dedicated to gauge the online influence of Twitter users based on a mixture of structural and interactional features. The model is an additive mathematical formulation which involves two main parts. The first part serves to measure the influence of the Twitter user on just his neighbourhood covering his followers. However, the second part evaluates the potential influence of the Twitter user beyond the circle of his followers. Particularly, it measures the likelihood that the tweets of the Twitter user will spread further within the social graph through the retweeting process. The model is tested on a data set involving four kinds of real-world egocentric networks. The empirical results reveal that an active ordinary user is more prominent than a non-active celebrity one. A simple comparison is conducted between the proposed model and two existing simplistic approaches. The results show that our model generates the most realistic influence scores due to its dealing with both explicit (structural and interactional) and implicit features.

Citation

Samir Akhrouf , Hamza Loucif, , Abdelhak Boubetra, , (2016), A simplistic model for identifying prominent web users in directed multiplex social networks: a case study using Twitter networks, New Review of Hypermedia and Multimedia, Vol:22, Issue:4, pages:287-302, Taylor and Francis online

One of the biggest challenges teachers are facing is the inability of students to interact with each other. The instructor may ask for students' opinion, their ideas and whether they agree or disagree on a given problem or statement, without getting any feedback. However, some students have communication difficulties, and those who can have the correct answer may feel uncomfortable, and therefore the class will miss a good learning opportunity. Our objective is to create real-time multiple choice question system with an interactive and fun learning environment to increase the productivity of the maximum number of students.

Citation

Samir Akhrouf , Mebarkia Mohamed Khalil, Cap Clemens H, Belayadi Yahia, Boubetra Djamel, ,(2016), A modern classroom response system for Bordj Bou Arreridj University,IMCL 2016,San Diego USA

Accurate singular point (SP) detection is an important factor in fingerprint (FP) recognition systems. We propose an algorithm to detect SPs in FP images. Our idea is based on the observation that the orientation field (OF) at the regions containing SPs has high variation, whereas in the other regions, it is smooth. Thus, a pixel-wise descriptor that comprises orientation-deviation (OD)-based features is proposed to measure the OF variation in the local neighborhood of a pixel which we call OF energy. Candidate SPs are characterized by locations where the OF energy function has local gradual maxima. Furthermore, the OD-based descriptor exhibits some advanced topological properties, in particular the descriptor profile tendency, which are highly correlated with the SP type. These properties are used to filter out some spurious SPs. A second refining step based on an extended Poincaré index is then applied to keep only genuine SPs with their information. The proposed algorithm has the ability to accurately detect the classical singularities as well as the arch-type SP. Experiments conducted over the public databases FVC2002 db1 and db2 confirm its accuracy and reliability with a reduced false alarm rate in comparison to other proposed methods.

Citation

Samir Akhrouf , Belhadj Foudil, Harous Saad, Ait Aoudia Samy, , (2015), Efficient fingerprint singular points detection algorithm using orientation-deviation features, Journal of Electronic Imaging, Vol:24, Issue:3, pages:13, SPIE Digital Library

Most of the solutions proposed to the security and privacy issues in mobile learning systems are generally inherited from those developed in e-learning systems and network communication background. However, the implication of the mobile devices in m-learning can reveal more private data and requires far more challenges. In this paper, we exploit the richness of mobile devices in sensors to propose a fingerprint-based strategy that provides a secure authentication, communication and non-repudiation scheme for mobile learning using recent advances in cancelable biometrics. The proposed scheme covers all the learning system steps starting b y subscription, communication and assessments.

Citation

Samir Akhrouf , Belhadj Foudil, Ait Aoudia Samy, ,(2015), Secure Fingerprint-based authentication and non-repudiation services for mobile learning systems,IMCL2015,Thessaloniki Greece

A significant development in virtual labs has been made in the last few years. The main reason for that is: the rapid development of technology, especially the internet, the huge growth of computer networks, the large use of personal computers and mobile devices. The World Wide Web with its different web services offers users a new way of learning; by giving the possibility of sharing and managing resources online and performing experiments through the web. This paper describes the use of BBA1 virtual laboratory through mobile devices. It will introduce the different tools available and will present some of the virtual experiments by giving an example of its usage in the virtual mobile laboratory.

Citation

Samir Akhrouf , Adel Merabet, Djamel Boubetra, Foudil Belhadj, Selmani Larbi, Abdelhak Boubetra, ,(2015), BBA virtual laboratory through m-learning,IMCL 2015,Thessaloniki Greece

The aim of this paper is to develop an Intelligent Learning Object Repository for computer science educational institutions at first glance, in order to facilitate the accessibility to different Learning Objects and therefore, enhance their reusability, interoperability, while taking advantage of their granularity/aggregation features. This LOR will give teachers the possibility to design courses that may include one or many LOs, second, the courses that depends on each other are linked together as nodes to form a graph that represents the global learning experience. Finally, it is to integrate a recommendation system which recommends to students the best paths to follow during a learning experience using the multi-agents approach where the recommender agent uses the A star pathfinding technique.

Citation

Samir Akhrouf , Abdelbasset Rouabah, Larbi Selmani, Yahia Belayadi, ,(2015), Dihya: an Intelligent Learning Object Repository,IMCL 2015,Thessaloniki Greece

E- learning is gaining more popularity due to the technological development and the personalization it offers for learners and organizations. Most e-learning platforms support group work and collaboration between learners, which is a desirable educational practice. Thus, virtual team members must be flexible to achieve their goals in a cooperative environment. In order to improve learners experiences, and to encourage collaboration between them, we propose and validate a theoretical model that predicts the effect of trust on learners' forgiveness when a conflict occurs in a virtual team.

Citation

Samir Akhrouf , Meriem Laifa, ; Roya Imani Giglou, Ramdane Maamri, ,(2015), Trust and Forgiveness in Virtual Teams: A Study in Algerian E-Learning Context,IMCL 2015,Thessaloniki Greece

We live in an interdisciplinary world, and we participate in an interplay between society and technology. In online settings, trust is required to back up successful interactions and to filter the overflow of information. However, trust is dynamic and very sensitive; it can be strengthened by positive experiences and destroyed by negative ones. Thus, we chose forgiveness as a way to repair broken trust relationships in computer-mediated communications. We believe that forgiveness benefits can improve online users' experiences by repairing broken relationships and encourage cooperation. In this paper, we present an overview about forgiveness research and we hope to put the first basic stones to drive more attention to this complicated concept in the digital environment.

Citation

Samir Akhrouf , Laifa Meriem, Maamri Ramdane, ,(2015), An Overview of Forgiveness in the Digital Environment,IPAC 2015,Batna Algérie

There is a wealth of information created every day through computer-mediated communications. Trust is an important component to sustain successful interactions and to filter the overflow of information. The concept of trust is widely used in computer science in various contexts and for different aims. This variety can confuse or mislead new researchers who are interested in trust concept but not familiar enough with it to find relevant related work to their projects. Therefore, we give in this paper an overview of online trust by focusing on its social aspect, and we classify important reviewed work in an attempt to guide new researchers in this domain and facilitate the first steps of their research projects. Based on previous trust surveys, we considered the following criteria: (1) trust dimension and its research purpose, (2) the trusted context and (3) the application domain in which trust is applied.

Citation

Samir Akhrouf , Laifa Meriem, Maamri Ramdane, ,(2015), Online Social Trust: an Overview,IPAC 2015,Batna Algérie

Great development in distance learning has been made in the last two decades facilitated by the rapid evolution of technology, the huge development of computer networks and the large use of internet and personal computers. The World Wide Web with its different web services offers learners and teachers’ new ways of learning and teaching; students can study at their own pace and perform practical tasks remotely. We, at the University of Bordj Bou Arreridj, are involved in the eSience (rESeau maghrébIn de laboratoirEs à distance) project which aims to link three elabs around the Maghreb countries namely Algeria, Tunisia and Morocco. The project is coordinated by the University of Bordeaux 1, France with the participation of many European partners. These elabs will propose practical and theoretical teaching units that will be part of their degree program. In this paper we present our university’s platform architecture, focusing on the web interface. First we introduce the service broker, which allows students to communicate and use the various platforms of the network. Second we present the simulation server and give an example of its usage.

Citation

Samir Akhrouf , Nasser Eddine Mouhoub , Adel Merabet, Ayoub Maza, Djamel Boubetra, Larbi Selmani, Abdelhak Boubetra, , (2014), Web Services for Virtual Simulation, International Journal of Online and Biomedical Engineering, Vol:10, Issue:5, pages:9-11, IJOE

Nowadays distant learning is gaining more and more popularity due to the technological development and the flexibility it offers for the students. All around the world we see universities conceiving their distant learning platforms in order to offer to their students this service and help them study at their own pace. In this paper we describe BBA lab a draft of the distant lab of Bordj Bou Arreridj University. It is dedicated mainly to students in the field of electrical engineering and computer science. Bordj Bou Arreridj University is involved in a project which is financed by the Eurepean Community and aims to link three distant Elabs, the Algerian Elab, developed by Constantine 1 and Bordj Bou Arreridj Universities, the Moroccan and the Tunisian Elabs.

Citation

Samir Akhrouf , Djamel Boubetra, Jean Jacques Charlot, Larbi Slimani, Boubetra Abdelhak, Adel Merabet, , (2014), BBA Lab a Virtual Laboratory for Distant Learning, International Journal of Information and Education Technology, Vol:4, Issue:1, pages:39-41, IJIET

Determining the identity of a person automatically is still an actual problem. There is an urgent need to have automatic and reliable authentication systems in order to combat fraud and meet the requirements for different areas ranging from international border stations to accessing personal information. Biometrics is an alternative to the old methods of identification. It involves identifying persons from their physical or behavioral characteristics. Face, fingerprint, iris, etc, are examples of physical characteristics. This course aims to introduce first basic concepts related to the recognition of individuals, based on their physiological or behavioral biometric characteristics, to students and allow them to study the key techniques to achieve reliable systems for biometric recognition. In This paper we will present the architecture of the course which will be held on a virtual laboratory. The course objective is to give the student different skills which are necessary in this field of study. The course is divided in several parts; each one will present a specific topic and will provide different notions in order to let students understand theoretical concepts by practicing.

Citation

Samir Akhrouf , Larbi Selmani, Djamel Boubetra, Mostefai Messaoud, , (2014), A Biometric Identification Course for eSience Virtual Lab, Lecture Notes on Software Engineering, Vol:2, Issue:, pages:87-89, LNSE

Towards Content Based Image Retrieval with Personalized Semantic Spaces

Citation

Samir Akhrouf , Tekkali Aicha, Rima Merrouche, Anas Y Boubas, Saad Harous, ,(2014), Towards Content Based Image Retrieval with Personalized Semantic Spaces,International Conference on Computer Vision and Image Analysis, ICCVIA’2014,Ras Al Khaimah UNITED ARAB EMIRATES.

E-learning is becoming more and more used by students and teachers in higher education. It is a new way of performing practical works and acquiring knowledge in many fields, especially in technology. Gathering ressourses and sharing their usage is very beneficial for educational institutions and will reduce the cost of training. In this paper we present the architecture of a platform we are developing in our university for the purpose of distance learning. This platform is a part of a TEMPUS project for linking three e-labs situated in three maghrebian countries, namely Algeria, Morocco and Tunisia. The platform is mainly dedicated to our students and can be accessed remotely via internet. We will present its architecture, focusing on the web interface and we will see that we use web services for access control and for simulations. The last will be performed by tools such as AMS and TCAD from Silvaco.

Citation

Samir Akhrouf , Adel Merabet, Djamel Boubetra, Larbi Selmani, Abdelhak Boubetra, ,(2014), Virtual Simulation on the Web,IAJC-ISAM International Conference,Orlando USA

Great development in distance learning has beenmade in the two last decade; the principal reason for that isdue to the rapid evolution of technology, the huge develop-ment of computer networks, the large use of internet andpersonal computers. The World Wide Web with its differentweb services offers learners and teachers’ new ways oflearning and teaching; students can study at their own paceand perform practical works remotely.We, at the University of Bordj Bou Arreridj, are involved ina TEMPUS project; its acronym is eSience which stands forrESeau maghrébIn de laboratoirEs à distaNCE, it aims tolink three elabs around the Maghreb countries namely Al-geria, Tunisia and Morocco. The project is coordinated bythe University of Bordeaux 1, France with the participationof many other European partners. These elabs will proposepractical and theoretical teaching units to their students andwill be part of their degree program.In this paper we present the architecture of the platformproposed by our university and we will focus on the webinterface which permits to students to perform their practi-cal works on the virtual simulator called SMARTSPICE ofSILVACO. The first part of the paper is an introduction tothe service broker, which will allow students from differentuniversities (partners) to communicate and use their differ-ent platforms without having to enroll in the different plat-forms. In the second part we present the simulation serverand give an example of its usage

Citation

Nasser Eddine Mouhoub , Samir Akhrouf , ,(2014), Web Services for Virtual Simulation,International Conference on Remote Engineering and Virtual Instrumentation (REV2014),Porto, Portugal

Social media and Social Network Analysis (SNA) acquired a huge popularity and represent one of the most important social and computer science phenomena of recent years. One of the most studied problems in this research area is influence and information propagation. The aim of this paper is to analyze the information diffusion process and predict the influence (represented by the rate of infected nodes at the end of the diffusion process) of an initial set of nodes in two networks: Flickr user’s contacts and YouTube videos users commenting these videos. These networks are dissimilar in their structure (size, type, diameter, density, components), and the type of the relationships (explicit relationship represented by the contacts links, and implicit relationship created by commenting on videos), they are extracted using NodeXL tool. Three models are used for modeling the dissemination process: Linear Threshold Model (LTM), Independent Cascade Model (ICM) and an extension of this last called Weighted Cascade Model (WCM). Networks metrics and visualization were manipulated by NodeXL as well. Experiments results show that the structure of the network affect the diffusion process directly. Unlike results given in the blog world networks, the information can spread farther through explicit connections than through implicit relations.

Citation

Samir Akhrouf , Nasser Eddine Mouhoub , Laifa Meriem, Belayadi Yahia, ,(2013), Social Network Analysis and Information Propagation,International Journal of Future Computer and Communication,Malaysia

Several authors have shown that multimodality was a highly efficient solution for biometric identification and allows to increase performance and reliability. Indeed, the acquisition of multiple biometric features makes it much more difficult for an impostor to spoof the system with artificially developed samples. However, we do not get these benefits for free: multimodal biometric systems are less profitable and have significant effects on their users. Some of them may indeed lead a non acceptance by their users, in particular when it comes to privacy issues that may result when acquiring data at multiple levels. That is why it is always important to explore the possibility of finding new systems that could be accepted by users and which are fully reliable than the previous ones. In this paper we present a new approach for speaker biometric identification. This approach is a user friendly platform which adds a new modality, namely a virtual character, which will be easily accepted by users. The virtual character [6] will first put the users in a comfortable position and simultaneously acquire the information necessary for their identification. Then, it will guide them step by step until the end of the process of identification. The users are put in the best conditions which will facilitate their participation and acceptance to be identified.

Citation

Samir Akhrouf , Nasser Eddine Mouhoub , Belayadi Yahia, , (2012), A New multimodal Biometric identification approach, International Journal of Computer Science Issues, Vol:9, Issue:2, pages:238-240, IJCSI

A scheduling problem is organizing in time a set of activities, so as to satisfy a set of constraints and optimize the result. The temporal constraint modifies the project scheduling, therefore in loses its characteristics. Our objective is to solve this problem by finding the various types of temporal constraints then modeling them by using graphs. Furthermore we apply a technique for transforming an AoN graph (Activities on Nodes) which is unique and contains a significant number of arcs. This graph is not preferred by practitioners of project management. We transform the AoN graph into an AoA graph (Activities on Arcs) which contains fewer arcs and is preferred by practitioners of project management. In this paper we present some concepts of line graphs and an illustrative example of the proposed method.

Citation

Nasser Eddine Mouhoub , Samir Akhrouf , , (2012), Generating Pert Network with Temporal Constraints, Studia Univ. BABES-BOLYAI, INFORMATICA, Vol:57, Issue:4, pages:03-18, Studia University BABES-BOLYAI, INFORMATICA

Le problème d’ordonnancement consiste à organiser dans le temps un ensemble d’activités, de façon à satisfaire un ensemble de contraintes et optimiser le résultat. La contrainte temporelle touche au problème central de l’ordonnancement et le modifie, donc il n’aura plus ses caractéristiques. Notre travail consiste, pour résoudre ce problème, à modéliser par les graphes ce type de contraintes et le ramener à un problème classique d’ordonnancement ; ensuite appliquer une nouvelle technique de transformation d’un graphe AoN (Activities on Nodes) qui est unique avec un nombre important d’arcs et non préféré par les praticiens de gestion de projet vers un graphe AoA (Activities on Arcs) avec moins d’arcs et plus souhaitable. Le papier contient des concepts sur les graphes adjoints et un exemple illustratif de la méthode proposée.

Citation

Nasser Eddine Mouhoub , Samir Akhrouf , Benhocine Abdelhamid, ,(2012), Construction et génération d’un graphe AoA en tenant compte des contraintes temporelles,Colloque sur l'Optimisation et les Systèmes d'Information COSI'2012,Tlemcen Algérie

Social media and Social Network Analysis (SNA) acquired a huge popularity and represent one of the most important social and computer science phenomena of recent years. One of the most studied problems in this research area is influence and information propagation. The aim of this paper is to analyze the information diffusion process and predict the influence (represented by the rate of infected nodes at the end of the diffusion process) of an initial set of nodes in two networks: Flickr user’s contacts and YouTube videos users commenting these videos. These networks are dissimilar in their structure (size, type, diameter, density, components), and the type of the relationships (explicit relationship represented by the contacts links, and implicit relationship created by commenting on videos), they are extracted using NodeXL tool. Three models are used for modeling the dissemination process: Linear Threshold Model (LTM), Independent Cascade Model (ICM) and an extension of this last called Weighted Cascade Model (WCM). Networks metrics and visualization were manipulated by NodeXL as well. Experiments results show that the structure of the network affect the diffusion process directly. Unlike results given in the blog world networks, the information can spread farther through explicit connections than through implicit relations.

Citation

Nasser Eddine Mouhoub , Samir Akhrouf , ,(2012), Social Network Analysis and Information propagation: A case study using Flickr and YouTube networks,International Conference on Information and Multimedia Technology (ICIMT 2012),Kuala Lumpur, Malaysia

Le problème d’ordonnancement consiste à organiser dans le temps un ensemble d’activités, de façon à satisfaire un ensemble de contraintes et optimiser le résultat. La contrainte temporelle touche au problème central de l’ordonnancement et le modifie, donc il n’aura plus ses caractéristiques. Notre travail consiste, pour résoudre ce problème, à modéliser par les graphes ce type de contraintes et le ramener à un problème classique d’ordonnancement ; ensuite appliquer une nouvelle technique de transformation d’un graphe AoN (Activities on Nodes) qui est unique avec un nombre important d’arcs et non préféré par les praticiens de gestion de projet vers un graphe AoA (Activities on Arcs) avec moins d’arcs et plus souhaitable. Le papier contient des concepts sur les graphes adjoints et un exemple illustratif de la méthode proposée.

Citation

Nasser Eddine Mouhoub , Samir Akhrouf , Abdelhamid Benhocine, ,(2012), Construction et génération d’un graphe AoA en tenant compte des contraintes temporelles,Colloque international sur l’optimisation et les systèmes d’information COSI’2012,Tlemcen Algérie

Nowadays Person Recognition has got more and more interest especially for security reasons. The recognition performed by a biometric system using a single modality tends to be less performing due to sensor data, restricted degrees of freedom and unacceptable error rates. To alleviate some of these problems we use multimodal biometric systems which provide better recognition results. By combining different modalities, such us speech, face, fingerprint, etc., we increase the performance of recognition systems. In this paper, we study the fusion of speech and face in a recognition system for taking a final decision (i.e., accept or reject identity claim). We evaluate the performance of each system differently then we fuse the results and compare the performances.

Citation

Samir Akhrouf , Belayadi Yahia, Mostefai Messaoud, Chahir Youssef, , (2011), A Multi-Modal Recognition System Using Face and Speech, International Journal of Computer Science Issues, Vol:8, Issue:3, pages:230-236, IJCSI

In real applications of speaker verification [1] based on GMM-UBM method, the size of training data varies from a speaker to another; the choice of a common threshold of verification will reduce the performance of the system when we only have short records of training data. In this paper we first describe speaker verification using adapted gaussian mixture models based on Reynolds works [2] and the performance obtained by this method. We then propose a new method to estimate a speaker-dependent threshold and the results obtained by this approach.

Citation

Samir Akhrouf , Mehemel Bachir, Mehemmel Abbas, ,(2011), Speaker dependent threshold’s estimation for GMM-UBM speaker verification systems,International Conference on Software and Information Management,Chengdu, China

The goal of this project is to bring together and integrates the work of the laboratory team members in order to get a practical realization of an Intelligent Multimodal Biometric Identification System. This last exploits the vocal and visual properties of a person to carry out, her or his, identification. To improve classification performance, the system will have a virtual character module which can exchange information with the person to be identified according to a random survey (based on information provided previously). Our first objective is to equip our laboratory with a powerful access control system able to identify the members of the lab and follows them where ever they go helping them with all kind of information they may be in need to.

Citation

Samir Akhrouf , Abderraouf. Bouziane, A. Hacine. Gharbi, Messaoud Mostefai, Youssef Chahir, , (2010), Towards an Intelligent Multimodal Biometric Identification System, International Journal of Computer Science and Electrical Engineering, Vol:2, Issue:6, pages:1001-1004, IACSIT

Développement d'un système d'identification biométrique multimodale

Citation

Samir Akhrouf , Bouziane Abderraouf, Hacine Gharbi Abdennour, Mostefai Messaoud, ,(2009), Développement d'un système d'identification biométrique multimodale,ICAI09,Bordj Bou Arreridj Algérie

A Study of an Online Auction Implementation

Citation

Samir Akhrouf , Messaoud Mostefai, Bouziane Abderraouf, ,(2009), A Study of an Online Auction Implementation,ICAI09,Bordj Bou Arreridj Algérie

Interpolation Intercoupe Markovienne pour la Reconnaissance 3D des Images Médicales Anisotropiques.

Citation

Samir Akhrouf , Belhadj Foudil, Ait Aoudia Samy, ,(2009), Interpolation Intercoupe Markovienne pour la Reconnaissance 3D des Images Médicales Anisotropiques.,ICAI09,Bordj Bou Arreridj Algérie

J2EE : Etude de cas d’une Application Web 3 tiers.

Citation

Samir Akhrouf , Benziouche Hamza, ,(2009), J2EE : Etude de cas d’une Application Web 3 tiers.,ICAI09,Bordj Bou Arreridj Algérie

Research in the field of face recognition knew considerable progress during these last years. Among the most evoked techniques we find those which employ the optimization of the size of the data in order to get a representation which makes it possible to carry out the recognition. For these methods, the images of faces are seen like points in a space of very great dimensions. The basic idea is to encode the initial data to pass to another space of dimensions much more reduced while preserving as much useful information. This paper presents a hybrid method combining principal components analysis (PCA) and the discrete cosine transform (DCT).

Citation

Samir Akhrouf , Sehili Mohamed Amine, Chekhchoukh Abdeslam, Chahir Youssef, Mostefai Messaoud, ,(2009), Face Recognition Using PCA and DCT,2009 Fifth International Conference on MEMS NANO, and Smart Systems,Dubai UAE

Development of A New Face tracking Operator

Citation

Samir Akhrouf , Mostefai Messaoud, Bouziane Abderraouf, ,(2008), Development of A New Face tracking Operator,ICEE08,Oran Algérie

Effecient Face Location Operator based on motion detection techniques

Citation

Samir Akhrouf , Mostefai Messaoud, Bouziane Abderraouf, ,(2008), Effecient Face Location Operator based on motion detection techniques,2nd International Conference on Electrical Engineering Design and Technologies,Hammamet Tunisia

CMMS*net: A distributed and object-oriented Data Manipulation Language using CORBA

Citation

Samir Akhrouf , Bouziane Abderraouf, Mostefai Messaoud, ,(2008), CMMS*net: A distributed and object-oriented Data Manipulation Language using CORBA,2nd International Conference on Electrical Engineering Design and Technologies,Hammamet Tunisia

The proposed approach seeks to develop a combined simulation tool based on Petri nets, the producer consumer computing model and the simulation techniques. This tool is designated for simulating manufacturing systems. Such an approach offers an adequate platform to carry out simulation in parallel to take advantage of that, plus the experts of the Petri nets will find a strategy to make simulations not far from their concepts of modelling.

Citation

Samir Akhrouf , Boubetra Abdelhak, Belayadi Yahia, , (2007), A specification tool for simulating a manufacturing system component, Journal of Advances in Computer Science and Engineering, Vol:1, Issue:1, pages:1-10, Pushpa Publishing House

In this communication we will present the different methods used for free form deformation over different types of models.

Citation

Samir Akhrouf , Nasser Eddine Mouhoub , Boubetra Adelhak, Fares Nour El Houda, ,(2007), A Study of Free Form Deformations and their Applications,Conférence Internationale sur la productique,Université Ferhat Abbas, Sétif

Les Graphes Adjoints dans le Problème Centrale d'Ordonnancement

Citation

Nasser Eddine Mouhoub , Samir Akhrouf , Boubetra Abdelhak, ,(2007), Les Graphes Adjoints dans le Problème Centrale d'Ordonnancement,Colloque International d'Analyse non Linéaire et Applications,Université de Sétif Algérie

Inside a descrete event simulation

Citation

Samir Akhrouf , Nasser Eddine Mouhoub , Abdelhak Boubetra, ,(2005), Inside a descrete event simulation,Congrès International en Informatique Appliquée,Bordj Bou Arreridj Algérie