Département d'informatique
Permanent URI for this communityhttp://dspace.univ-batna2.dz/handle/123456789/6
Browse
133 results
Search Results
Item Partage de Secret en Utilisant des Métaheuristiques Bionispirées(University of Batna 2, 2024) ZENDER RouiaTechnology has progressed a lot in recent decades, especially in terms of computing power and will probably continue to do so in the future, Moore's law is there to support it. This had disastrous repercussions on the information security; although cryptography has also evolved a lot in terms of encryption algorithms, recent supercomputers are capable of overcoming some cryptosystems. This has led researchers in this field to investigate new avenues, such as secret sharing, which over time has proven to be the most secure solution currently used for protecting its secret. The main axis around which revolves the subject of this thesis is precisely this secret sharing paradigm, requiring cooperation and collaboration on the part of secret keepers based on distributed trust and jointly managed control of the security situation.During these last years, various sharing algorithms have been proposed. Indeed, security in the field of IT in general and computer systems in particular have contributed a lot in the evolution of this discipline which belongs to cryptography and which is based on strong theoretical pillars such as: modular arithmetic and coding theory. Contribution in this research work consists of the proposal of a new secret sharing scheme based on bioinspired hexagonal structure and integer decomposition. As a metaheuristic inspiration, hexagonal structure was inspired from nature where it’s common in constructs made by biological systems and the intelligent behaviours of a bee swarm. For integer decomposition, it is known that the oldest method is Fermat’s factorization, which is based on the representation of an odd integer as the difference of two squares, while for the proposed decomposition every positive integer has a unique factorization into two factors. To check functionality and efficiency of the proposed scheme, it was applied to digital images processing domain where it has exhibited good properties: it is lossless and ideal, its flexibility allows many extensions to handle additional situations, it can add new or delete an old participant and it can detect and identify cheater. Aside these interesting features, experimental results demonstrate that this scheme has a good security level.Item Un environnement d'exécution de simulation basé sur le calcul volontaire(University of Batna 2, 2020) KADACHE NABILLe calcul volontaire (CV) est devenu une technique relativement mature de calcul distribué. Son principe consisté a exploiter le temps de repos des machines ordinaires connectées a internet et avec le consentement de leurs propriétaires. Les applications cibles sont généralement des projets scientiques nécessitant un temps et des ressources énormes de calcul. Les plateformes de calcul volontaire existantes soulévent plusieurs dés concernant les diérentes fonctions qui doivent ^etre assurées. Dans une premére partie de notre travail, nous essayons d'apporter des solutions a deux dés inhérents au CV ; Le premier concerne l'implication de volontaires, qui constitue le maillon faible de ce type de systémes. Nous proposons un nouvel environnement de calcul volontaire social (SVCE) intégrant les fonctionnalités récentes de Facebook afin d'impliquer davantage de volontaires. Le deuxiéme probléme, que nous avons abordé, est celui de l'ordonnancement des t^aches dans les systémes de CV. Pour cela, nous proposons un algorithme qui consistéa générer, pour chaque volontaire, un nombre de taches élémentaires dont le cout d'exécutionre éte la capacité de calcul momentanée des ressources disponibles des volontaires, la validité de notre algorithme est illustrée expérimentalement. Dans la deuxiéme partie de notre thése, nous nous penchons sur la simulation distribuée, nous proposons à cet éet un environnement de simulation (VolSIM) qui combine les techniques liées au calcul volontaire àn d'exécuter des simulations qui requiérent des ressources conséquentes.Item Machine Learning Pour La Détection Des Communautés(University of Batna 2, 2024) Wafa LOUAFIThe burgeoning field of social network analysis has garnered considerable attention, with a particular focus on the critical area of community detection. Communities, construed as clusters of closely interconnected nodes with weaker ties to the broader network, play a pivotal role in understanding the evolution of network structures, an essential aspect of network analysis. This thesis conducts a comprehensive exploration into community detection algorithms and unsupervised learning methods, subsequently delving into the application of machine learning techniques in this domain. Two distinct methods are presented: one focused on detecting overlapping communities and the other on identifying disjoint communities. Notably, our methods involve the selection of vital nodes through subgraphs, operating individually on each node, while the clustering process itself occurs globally across the entire network. The implementation leverages various unsupervised machine learning techniques, including hierarchical clustering and k-means, showcasing efficiency and ease of implementation. The standout feature of our methods lies in their demonstrated superiority, achieving enhanced accuracy and performance compared to contemporary methodologies. Substantiating this claim, meticulous evaluations were conducted on both real and synthetic network datasets. Beyond these achievements, the research opens avenues for future exploration in understanding the broader implications and applications of community detection within evolving network structures.Item Automatic multi-documents text summarization using Binary Biology Migration Algorithm(University of Batna 2, 2024) Mohamed BOUSSALEMAs the World Wide Web continues to expand, the process of identifying pertinent information within its vast volume of documents becomes increasingly challenging. This complexity necessitates the development of efficient solutions, one of which is automatic text summarization; an active research area dedicated to extracting key information from extensive text. The difficulties are further compounded when addressing multi-document text summarization, due to the diversity of topics and sheer volume of information. In response to this issue, this study introduces a novel approach based on swarm intelligence algorithm called biology migration algorithm (BMA). Our proposed approach is; Binary Biology Migration Algorithm for Multi-Document Summarization (BBMA-MDS). Viewing multidocument summarization as a combinatorial optimization problem, this approach leverages the biology migration algorithm to select an optimal combination of sentences. Evaluations of the proposed algorithm's performance are conducted using the ROUGE metrics, which facilitate a comparison between the automatically generated summary and the reference summary, commonly known as the 'gold standard summary'. For a comprehensive evaluation, the well-established DUC2002 and DUC2004 datasets are employed. The results demonstrate the superior performance of the BBMA-MDS approach when compared to alternative algorithms, including firefly and particle swarm optimization, as indicated by the selected metrics. This study thus contributes effectively according to the evaluation to the field by proposing BBMA-MDS as an effective solution for the multi-document text summarization problem.Item Une approche coopérative basée sur l’IoT pour améliorer la qualité du trafic routier(University of Batna 2, 12/21/2023) Tarek Amine HADDADNowadays, automobiles have become very useful for the daily transportation of both people and goods. However, the increase in their numbers has generated a significant rise in demands for the use of road networks, which can lead to delays, traffic congestion, and poor traffic flow, especially in large cities and global metropolises. This type of problem often referred to as road congestion, can be caused by other factors such as road works, accidents, insufficient traffic management, etc. Many strategies to reduce traffic congestion have been adopted by governments and transport agencies. These strategies include building new roads, planning additional public transport routes, real- time traffic management, etc. On the other hand, traffic signal control (TSC) represents one of the modern trends that can play a crucial role in traffic management. It also refers to the management and coordination of traffic lights on road networks in order to effectively control traffic and improve traffic flow. This can involve adjusting the traffic light cycle time based on real traffic to optimize certain performance parameters. Modern TSC systems can adopt technologies such as traffic sensors, surveillance cameras and data processing algorithms to optimize traffic management. Reinforcement learning (RL) is an intelligent approach widely adopted by adaptive TSC systems to optimize traffic signal management, enhancing traffic flow. Indeed, it is possible to train a system to learn how to adjust traffic light cycle times based on the state of real traffic in order to reduce road congestion. In this thesis, we propose several cooperative approaches based on Deep Reinforcement Learning (DRL) to intelligently optimize the management of traffic lights in a road network with multiple intersections. We have thus modeled our problem as a multi-agent reinforcement learning system (MARL). This involves the use of multiple agents each of whom can learn to make decisions in terms of adjusting light cycle times to the local traffic situation, and these decisions can be synchronized with the decisions made by other agents to ensure optimal functioning of the entire system. In such approaches, each agent can receive from its neighbors their states, actions and rewards, combining them with its own state, action and reward to make the appropriate decisions. Experimental results under different scenarios show that the proposed approaches outperform many state-of-the-art approaches in terms of three parameters: Average Waiting Time (AWT), Average Queue Length (AQL) and average CO2 emission (AEC).Item Approche Multimodale basée sur l’Apprentissage Profond pour l’Identification Biométrique(University of Batna 2, 2023) ABDESSALAM HATTABDue to the growing need for user identification in various modern applications, experts highly recommend incorporating biometric technology in the application development field. Recently, many recognition systems using face and iris traits obtained remarkable performance, particularly since these biometric traits are captured from a distance without physical contact with sensors. This feature reduces the potential spread of the COVID-19 pandemic and other diseases by touch and makes biometrics more convenient and user-friendly. However, the recognition systems’ performance is significantly reduced when these traits are captured under uncontrolled conditions, including occlusion, poses and illumination variation. Because the handcrafted approach used by some recognition systems extracts local features from the global image, where the image’s regions affected by the uncontrolled conditions often influence the quality of extracted features. This thesis proposed a robust face recognition system that extracts features from important facial regions, using Scale-invariant feature transform (SIFT) to identify significant facial regions and adaptive Local Ternary Pattern (ALTP) to extract features. The proposed system achieved promising results on two benchmark face datasets captured under unconstrained conditions, achieving 99.75% on the ORL dataset and 95.12% on the FERET dataset. In addition, a new face recognition system based on Deep Learning has been proposed using the pre-trained AlexNet-v2 model. The proposed system achieved excellent results on the ORL and FERET face datasets, reaching 100% on the first and 99.89% on the second. Moreover, we proposed a novel face recognition system to address the issue of illumination variation. The system used a novel model inspired by the pre-trained VGG16 model. The proposed system achieved state-of-the-art results, reaching 99.32% on the Extended Yale B dataset and 99.79% on the AR dataset. Another contribution of this work is developing a novel iris recognition system based on Transfer Learning to achieve high accuracy rates. The proposed system used the Yolov4-tiny model to localize the iris region, while a novel Deep Convolutional Neural Network (CNN) model inspired by the pre-trained Inception-v3 model was used for features extraction. The performance of this system was evaluated on four different iris databases captured under non-cooperative conditions, where it achieved a new state-of-the-art accuracy rate reaching 99.91%, 99.60%, 99.91%, and 99.19% on the IITD, CASIA-Iris-v1, CASIA-Iris-Interval, and CASIA-Iris-Thousand, respectively. The proposed unimodal systems achieved high accuracy compared to state-of-the-art methods. However, relying solely on a unimodal biometric trait is inadequate for high-security requirements in military and government applications. Finally, three face-iris multimodal biometric systems have been proposed in this thesis. The first employs the fusion of images, the second utilizes feature-level fusion, and the third is based on score-level fusion. The proposed systems used Yolov4-tiny to detect the face and both iris regions. In addition, they applied a new deep CNN model inspired by the pre-trained Xception model to extract features. To evaluate the performance of the proposed systems, a twofold cross-validation protocol is employed on the CASIA-ORL and SDUMLA-HTM multimodal benchmark databases. The results showed that our systems achieved a perfect score of 100% on both databases. Remarkably, the system utilizing score-level fusion outperformed the other systems, achieving outstanding results of 100% on the CASIA-ORL database and over 99% on the SDUMLA-HTM database, with only one sample used for training.Item Numerical study of two-phase streams in a linear solar concentrator with a new S-shaped absorber. Application: Direct Steam Generation(University of Batna 2, 2023) ACHI AlladdineThis numerical study focuses on the feasibility of the new solar receiver proposed and patented by Demagh Yassine in 2015 for linear concentrating modules. After the promising numerical results obtained in single-phase flow with synthetic oil, it was interesting to move towards the Direct Steam Generation (DSG) in parabolic trough collectors by adopting the scenario where this absorber would replace the conventional straight one in a full-scale installation like the DISS test facility in Almeria, Spain. First, the mathematical model was validated by comparison with experimental data for boiling flows in straight and coiled horizontal tubes at a laboratory scale and followed by validation under actual conditions in the commercial scale of the DSG in the DISS facility. Preforming 3D simulations in the OpenFOAM environment, the obtained numerical results were more than satisfactory. In the second step, the previously validated model is used to perform simulations of DSG in the DISS facility parabolic trough collector equipped with the new S-curve absorber in replacement of the conventional straight one; the same (actual) operating conditions are adopted for both absorbers for better performance comparison. The curved shapes of the tube greatly disturb the stratified flow by converting it into a stratified wavy and slug flows, further improving the evaporative heat transfer coefficient. It is evident that a penalty accompanies any improvement in thermal performance in terms of pressure drops, but it remains entirely acceptable. The S-curved absorber is not recommended when the conventional straight absorber can provide an annular flow alone, which is not always true. Key words: Boiling flow, DSG, Parabolic trough, DISS test facility, S-shaped absorber, coiled tube, Monte Carlo Ray Tracing method.Item Approche d’apprentissage pour l’analyse des Big Data(University of Batna 2, 2023) Benoughidene AbdelhalimThe development of information technology has led to the big data revolution, with the amount of data produced increasing at a high rate. Video data is a significant component of big data, and the concept of automatically analyzing this rapidly increasing video content has become a popular research topic. In such a scenario, video automatic analysis uses a new generation of information technologies, such as artificial intelligence (machine learning), which help transform traditional video analysis to be more efficient and convenient. Video summary (VS) is now one of the primary areas of study in video analysis. Despite the use of big data-driven models, producing accurate video summaries in an efficient and effective manner remains a challenging task. The most effective and efficient way to transform lengthy, unstructured videos into structured, condensed, understandable, and useful information is through the use of video summaries. The primary goal of summarizing a video is to break it down into shots and select key frames from each shot that best capture the essence and flow of the entire video. The goal of this thesis is to improve the performance of video summarization systems by enhancing the quality of analytical techniques. To achieve this goal, we proposed two contributions. First, we suggested a shot boundary detection (SBD) method to adapt key shots and exploit its potential in video summaries. This is the first step in the video summarization process, and the results have a significant impact on the quality of the final summary. The main idea behind SBD is to extract features from video frames and then identify the boundaries between shots based on the differences in the features. Second, we focused on improving video summaries using unsupervised machine learning techniques (DBSCAN) and genetic algorithms (GA) to optimize DBSCAN hyperparameters. We validated the proposed methods and results obtained through extensive comparative analysis using datasets.Item Contrôle de qualité des architectures logicielles à base de composants : Cas d'étude.(University of Batna 2, 9/8/2022) KADRI SalimDepuis plusieurs décennies, la qualité des logiciels a été considérée comme le constituant clé du succès commercial des organisations. Les architectures logicielles ont été jugées comme le niveau approprié pour traiter les exigences de qualité des utilisateurs (attributs de qualité). Une multitude de méthodes a été proposée pour évaluer l'architecture logicielle. Cependant, selon l'étude de la littérature que nous avions effectuée, nous avons constaté que ces méthodes souffrent de nombreuse lacunes. Dans cette thèse, notre objectif est de surmonter ces lacunes en proposant une nouvelle méthodologie d'évaluation à l’aide d’un framework d'évaluation multiservice appelé MS-QuAAF. Ce framework se compose de deux modules principaux. Le premier module propose un nouveau concept appelé projection de facettes pour réduire la complexité d’une architecture cible en réduisant la taille de ses méta-modèles. Le deuxième module fournit un ensemble de métriques génériques appliquées aux architectures à travers trois services d'évaluation. Le premier service permet d'évaluer la défectuosité de l'architecture durant l’étape de conception. Le deuxième service est utilisé pour évaluer l'architecture à l’étape d’implémentation grâce à une méthode proposée appelée analyse de décomposition des responsabilités. Le troisième service est dédié à la finalisation de l'évaluation et à la production du rapport final. L'évaluation expérimentale que nous avions conduit à travers deux cas d’études nous a permis de répondre aux questions de recherche confrontées, ainsi de nombreuses conclusions et contributions ont été obtenues. Contrairement à la plupart des méthodes proposées, MS-QuAAF peut fournir une évaluation continue qui couvre deux étapes principales de développement : l’étape de conception et l’étape d’implémentation. Le framework permet à travers ses métriques génériques, l'évaluation de tous les attributs de qualité et la détection des déviations d’architecture. De plus, il est capable de guider les architectes pendant le processus de développement pour détecter les irrégularités et améliorer la qualité de l'architecture.Item Définition d’un style architectural pour la description de systèmes logiciels à base de composants de type COTS, selon une approche « services WEB ». Magister thesis, (2005) Université de Batna 2.(University of Batna 2, 5/31/2016) MANSOURI, KamelThe development of big software applications is oriented toward the integration or interoperation of existing software components (like COTS and legacy systems) . This tendency is accompanied by a certain number of drawbacks for which classical approaches in software composition cannot be applied and fail. COTS-based systems are built in ad-hoc manner and it is not possible to reason on them no more it is possible to demonstrate if such systems satisfy important properties like Quality Of Service and Quality Attributes. The recent works issued in web field allow the definition and the use of a complex web service architecture. Languages such as WSFL, XLANG and BPEL4WS support these architectures called Services Oriented Architectures. The definition of software systems using these languages benefits some existing technical solutions such as SOAP, UDDI, etc., that permit the distribution, the discovery and the interoperability of web services. However, these languages do not have any formal foundation. One cannot reason on such architectures expressed using such languages: properties cannot be expressed and the system dynamic evolution is not supported. On the other hand, software architecture domain aims at providing formal languages for the description of software systems allowing to check properties (formal analyses) and to reason about software architecture models. In this work, we proposes a formalisation of COTS-based system (their structure, their behaviours) using architectural styles. The ADL used is π- ADL (based on the π-calculus, supporting style description). The memory will also present our approach consisting in refining an abstract architecture to an executable and services-oriented one.