Lejdel Brahim

PHD student
Assistant professor
El-Oued University · Department of Computer science

Publications

  • Lejdel Brahima, Kazar Okbaa, Laurini Robertc
    [Show abstract] [Hide abstract]
    ABSTRACT: In a map, there are different relationships between spatial objects, such as topological, projective, distance, etc. Regarding topological relations, if the scale of the map is changed and if some spatial objects are generalized, not only the shapes of those objects will change (for instance, a small area becomes a point and then disappears as the scale diminishes), but also their topological relations can vary according to scale. In addition, a mathematical framework which models the variety of this category of relationships does not exist. In the first part of this paper, a new topological model is presented based on ribbons which are defined through a transformation of a longish rectangle; so, a narrow ribbon will mutate to a line and then will disappear. Suppose a road is running along a lake, at some scales, they both appear disjointed whereas at some smaller scales, they meet. So, the topological relations mutate according to scale. In this paper, the different components of this mathematical framework are discussed. For each situation, some assertions are defined which formulate the mutation of the topological relationships into other ones when downscaling.
    Journal of Visual Languages & Computing 11/2014; · 0.56 Impact Factor
  • Lejdel Brahim
    [Show abstract] [Hide abstract]
    ABSTRACT: : The generalization process of spatial data is a very important process for delivering to the user a map according to its needs. Many approaches were proposed for modeling and implementing this process. The system multi-agent is the most used. But, the central problem of this approach is the selection of the optimal action performed by the agent in a given moment. Several mechanisms are used for this class of problem. In this paper, we proposed a hybrid approach that can optimize actions of SMA with using the improved genetic algorithm. This new hybrid optimization algorithm combines the techniques of genetic algorithm and tabu search method (GA-TS). We introduced the TS into the genetic algorithm to ensure an efficient of the best solution. The objective is to minimize the time of generalization process and resolving all the conflicts constraints for improving the quality of generalized map.
    International Journal of Emerging Trends & Technology in Computer Science (IJETTCS). 07/2013; 2(2):155-160.
  • Lejdel Brahim, Kazar Okba
    [Show abstract] [Hide abstract]
    ABSTRACT: The creation of the map on-the-fly becomes today an important domain in cartography. Thousands of the users need access to spatial data on the web specific to their needs. This requires generating a map in real time. Different approaches appeared for optimizing the on-the-fly generation maps, but those not suffice for guiding a powerful and efficient process. Thus, a new approach must be developed for optimizing the on-the- fly generalization process according to the user needs. This paper describe a new approach that combines the multi agent system and the techniques of genetic algorithm which improved by Tabu search method (GA- TS). We introduced the TS into the genetic algorithm to refine the solutions space found by GA toward the optimum solution. This approach uses the multiple representation and cartographic generalization.
    The International Conference on Information Technology, IEEE 6th; 05/2013
  • Brahim LEJDEL, Okba Kazar
    International Journal of Computer and Information Technology. 11/2012; 1(2).
  • Brahim Lejdel, Kazar Okba
    [Show abstract] [Hide abstract]
    ABSTRACT: The generalization process of spatial data is a very important process for delivering to the user a map according to its needs. Many approaches were proposed for modeling and implementing this process. The system multi-agent is the most used. But, the central problem of this approach is the selection of the optimal action performed by the agent in a given moment. Several mechanisms are used for this class of problem. In this paper, we proposed a hybrid approach that can optimize actions of SMA with using the improved genetic algorithm. This new hybrid optimization algorithm combines the techniques of genetic algorithm and tabu search method (GA-TS). We introduced the TS into the genetic algorithm to ensure an efficient of the best solution. The objective is to minimize the time of generalization process and resolving the most important conflicts constraints for improving the quality of generalized map. Keywords: Generalization process, Agent, Genetic Algorithm, Tabu Search, spatial constraints. 1. INTRODUCTION Automatic generalization process has been an important research area in the current GIS and continues in the future. It is a set of operations, inspired of traditional cartographic generalization. Its main role is to simplify geographic data when they are very detailed, in order to satisfy user needs in cartographic applications. The principal objective of this process is creating a clear map from a geographic database vector very detailed but this process is very complex and very long. Several methods and concepts proposed to automate the generalization process but a framework for their combination into a comprehensive automatic generalization process is missing [1]. Several works model the spatial objects by agents such as the works of ([1] and [2], [13]). The strategy presented in [2] is the recent proposition for automated the generalization process; nevertheless, it is not flexible enough since it does not enable the agent to choose the best action to perform according to a given situation [20]. According to our idea, achieve optimal generalization process of spatial data implied that the agent performs at any time, an optimal plan which satisfy the different constraints. Thus, in artificial intelligent, choose an optimal action by agent is key mechanism for design an intelligent system. Optimal Action Selection allows to an agent, to determine at any instant what the optimal action that do in the next [16]. More formally, given an agent some knowledge of its internal state, and some sensory information concerning environmental context, permits it to decide what action (or action sequence) to perform in order to achieve its goals. There are two key questions in optimal action selection [10]:  What is being selected?  How is it being selected? In this paper, we propose to provide the agent a set of capacity; perception, communication and optimizer that allow it to select the optimal plan. In this context, optimal
    International Journal of Emerging trends & technologie in computer science. 08/2012; 2(2):155-160.
  • Brahim Lejdel, Okba Kazar
    [Show abstract] [Hide abstract]
    ABSTRACT: The utilization of web mapping becomes increasinglyimportant in the domain of cartography. Users want access to spatial data on the web specific to their needs. For this reason, different approaches were appeared for generating on-the-fly the maps demanded by users, but those not suffice for guide a flexible and efficient process. Thus, new approach must be developed for improving this process according to the user needs. This work focuses on defining a new strategy which improves on-the-fly map generalization process and resolves the spatial conflicts. This approach uses the multiple representation and cartographic generalization. The map generalizationprocess is based on the implementation of multi-agent system where each agent was equipped with a genetic patrimony.
    International Journal of Information Technology and Convergence and Services. 06/2012; 2(3):1-10.
  • Brahim lejdel
    [Show abstract] [Hide abstract]
    ABSTRACT: The automatic generalization processes consist to derive less detailed spatial data from data too detailed. The objective of this process is to provide the user a spatial data adapted to those needs. Several approaches are proposed to automate this process. These approaches include the agent-based approach. But the central problem of this approach is the selection of the optimal action performed by the agent in a given moment. In this paper, we proposed an approach that can optimize this process by satisfying the cartographic constraints. Our approach consist to provide agents geographic, genetic patrimony, to enable them choosing the optimal action, from where the concept of genetic agent.
    International Journal of Digital Information and Wireless Communications (IJDIWC). 01/2012; 3(1):729-737.
  • Lejdel Brahim
    [Show abstract] [Hide abstract]
    ABSTRACT: Modeling and managing multiple representations of the same real geographical phenomenon stocked in Multiple Representation Spatial Databases (MRSDB) has an important role in the new GIS, now and will continue in the future. The objective of these MRSDB is to serve different groups of users according to theirs domains applications, by accurate geographical information in suitable time. Many approaches have been developed for modeling the multiple representations of spatial data. They are based on multiple representations or generalization and taking into account, the two main parameters of multiplicity; level of detail and viewpoint. But the cost and complexity associated with modeling, management and automation of generalization are the main disadvantages of these approaches. In this paper, we propose a new approach, which have baptized “Hybrid approach for multiple representations”. It’s based firstly, on the storage of pertinent representations in MRSDB and partly on the generation of intermediates representations using a automatic generalization process, optimized by introducing a package of pre-classifying of data according to specific criteria and constraints that we have set a priori.
    International Conference on Information and Communication Systems (ICICS 2009); 12/2009

471 Following View all

115 Followers View all