-
2014Islem Henane, , ,
Towards a generic approach for multi-level modeling of renewable resources management systems
Proceedings of the 2014 International Conference on Autonomous Agents and Multi-Agent Systems, 1471–1472. Presented at the Paris, France. Richland, SC: International Foundation for Autonomous Agents and Multiagent Systems., 2014
Abstract
Multi-agent systems are widely used in renewable and natural resources management. Multi-agent systems are able to manage the complexity of such systems characterized by a large number of interacting entities with different levels of granularity and including dynamics of different contexts (ecological, economic, social). In this work, we propose a generic multi-level architecture for renewable and natural resources management.
Haithem MezniTowards trustworthy service adaptation: An ontology-based cross-layer approach
IEEE 5th International Conference on Software Engineering and Service Science, 2014
Abstract
Although several approaches have been proposed towards self-adaptation of Web services, most of them work in isolation and few of them deal with cross-layer and trust issues. Indeed, the complex layered nature of service-based systems frequently leads to service failure and conflicting adaptation. To tackle this problem, we propose an ontology-based categorization of service behavior across all the functional layers. The proposed ontology provides support for cross-layer self-adaptation by facilitating reasoning about events to identify the real source of service failure, and reasoning about self-adaptation actions to check integrity and compatibility of self-adaptation with constraints imposed by each layer.
Lamjed Ben Said, Zahra Kodia,Design Of Cognitive Investor Making Decision For An Artificial Stock Market Simulation: A Behavior-based Approach
Soft-Computing in Capital Market: Research and Methods of Computational Finance for Measuring Risk of Financial Instruments,2014, 41-56., 2014
Abstract
Computational Finance, an exciting new cross-disciplinary research area, depends extensively on the tools and techniques of computer science, statistics, information systems and financial economics for educating the next generation of financial researchers, analysts, risk managers, and financial information technology professionals. This new discipline, sometimes also referred to as « Financial Engineering » or « Quantitative Finance » needs professionals with extensive skills both in finance and mathematics along with specialization in computer science. Soft-Computing in Capital Market hopes to fulfill the need of applications of this offshoot of the technology by providing a diverse collection of cross-disciplinary research. This edited volume covers most of the recent, advanced research and practical areas in computational finance, starting from traditional fundamental analysis using algebraic and geometric tools to the logic of science to explore information from financial data without prejudice. Utilizing various methods, computational finance researchers aim to determine the financial risk with greater precision that certain financial instruments create. In this line of interest, twelve papers dealing with new techniques and/or novel applications related to computational intelligence, such as statistics, econometrics, neural- network, and various numerical algorithms are included in this volume.
-
2013, Wassim Ayadi, , ,
Survey on Biclustering of Gene Expression Data
Biological Knowledge Discovery Handbook: Preprocessing, Mining, and Postprocessing of Biological Data, 2013
Abstract
Microarrays allow measuring the expression level of a large number of genes under different experimental samples or environmental conditions. The data generated from them are called gene expression data. Gene expression data are usually represented by a matrix M, where the ith row represents the ith gene, the jth column represents the jth condition, and the cell mij represents the expression level of the th gene under the jth condition. In this chapter, the authors make a survey on biclustering of gene expression data. First, the chapter presents the different types of biclusters and groups of biclusters. Then, it discusses the evaluation functions and systematic and stochastic biclustering algorithms. Finally, the chapter focuses on bicluster validation that can qualitatively evaluate the capacity of an algorithm to extract meaningful biclusters from a biological point of view.
, Lamjed Ben SaidA DISTRIBUTED PRIVACY-PRESERVING MODEL FOR E-SERVICES
International Conference on Internet Technologies & Society., 2013
Abstract
In this paper, we propose a model for privacy protection of users in the context of e-services. A system based on our model has to respect a set of properties to preserve the user privacy. These properties are formulated as a set of privacy constraints: the anonymity, the pseudonymity, the unobservability and the unlinkability constraints. To satisfy these constraints we use the Distributed Constraint Satisfaction approach such that: (1) the variables correspond to the user’s credentials, (2) the agents correspond to the set of e-services entities that control these variables and (3) the constraints correspond to the set of privacy constraints. A solution to the problem is achieved when all the privacy constraints are satisfied. To validate the applicability of our proposed model, a set of experimentation results are discussed.
, , , Lamjed Ben SaidMulti-Agent System Model for Container Management Simulation.
In : ICEIS (1). 2013. p. 498-505., 2013
Abstract
This paper discusses an approach to build a multiagent system for simulating container management in a hub port logistics. The simulator has as goal to help assessing and defining container management strategies. This allows to plan and to control the management of containers while minimizing the waiting time and the parasite shifts and insuring the consistency of the performed tasks sequence. The proposed model involves the multipoint of view and the emergence of behavior specific to the theory of complex systems. The paper is structured as follows: first we present related works, then we expose the multiagent model of the simulator, after that we present the internal structure of the agents and finally we provide and discuss first implementation and results.
, Lamjed Ben Said,Multiobjective Analysis of the Multi-Location Newsvendor and Transshipment Models
International Journal of Information Systems and Supply Chain Management (IJISSCM), 6(4), 42-60., 2013
Abstract
Unlike the Newsvendor model, a system based on lateral transshipments allows the unsold inventories to be moved from locations with surplus inventory to fulfill more unmet demands at stocked out locations. Both models were thoroughly studied and researches were usually confined to cost minimization or profit maximization. In this paper, the authors proposed a more realistic multiobjective study of both multi-location Transshipment and Newsvendor inventory models. The aggregate cost, the fill rate, and the shared inventory quantity are formulated as conflicting objectives and solved using two reference multiobjective evolutionary algorithms (SPEA2 and NSGA-II). The proposed models take into account the presence of storage capacity constraints. The obtained Pareto fronts revealed interesting information. When transshipments are allowed, both low aggregate cost and high fill rate levels are ensured. The required shared inventory may have an important variability. The considered objective functions are conflicting and very sensitive to local storage capacities.
, Moez Hammami,Ensemble classifiers for drift detection and monitoring in dynamical Environments
Annual Conference of the Prognostics and Health Management Society 2013, 2013
Abstract
Detecting and monitoring changes during the learning process are important areas of research in many industrial applications. The challenging issue is how to diagnose and analyze these changes so that the accuracy of the learning model can be preserved. Recently, ensemble classifiers have achieved good results when dealing with concept drifts. This paper presents two ensembles learning algorithms BagEDIST and BoostEDIST, which respectively combine the Online Bagging and the Online Boosting with the drift detection method EDIST. EDIST is a new drift detection method which monitors the distance between two consecutive errors of classification. The idea behind this combination is to develop an ensemble learning algorithm which explicitly handles concept drifts by providing useful descriptions about location, speed and severity of drifts. Moreover, this paper presents a new drift diversity measure in order to study the diversity of base classifiers and see how they cope with concept drifts. From various experiments, this new measure has provided a clearer vision about the ensemble’s behavior when dealing with concept drifts.
, , Moez HammamiNouvelle méthode de détection de dérive basée sur la distance entre les erreurs de classification
5e Journées Doctorales Journées Nationales MACS, Strasbourg : France (2013), 2013
Abstract
La classification dynamique s’intéresse au traitement des données non-stationnaires issues des environnements évolutifs dans le temps. Ces données peuvent présenter des dérives, qui affectent la performance du modèle d’apprentissage initialement construit. Aujourd’hui, beaucoup d’intérêts sont portés sur la surveillance, la mise à jour et le diagnostic de ces dérives afin d’améliorer la performance du modèle d’apprentissage. Dans ce contexte, une nouvelle méthode de détection de dérive basée sur la distance entre les erreurs de classification est présentée. Cette méthode, nommée EDIST, surveille la distribution des distances des erreurs de classification entre deux fenêtres de données afin de détecter une différence à travers un test d’hypothèse statistique. EDIST a été testée à travers des bases de données artificielles et réelles. Des résultats encourageants ont été trouvés par rapport à des méthodes similaires. EDIST a pu trouver les meilleurs taux d’erreur de classification dans la plupart des cas et a montré une robustesse envers le bruit et les fausses alarmes.
, Meriem EnnigrouParticle Swarm Optimization combined with Tabu Search in a Multi-Agent model for Flexible Job Shop Problem
ICSI, 2013
Abstract
Flexible job shop scheduling problem (FJSP) is an important extension of the classical job shop scheduling problem, where the same operation could be processed on more than one machine and has a processing time depending on the machine used. The objective is to minimize the makespan, i.e., the total duration of the schedule. In this article, we propose a multi-agent model based on the hybridization of the tabu search (TS) method and particle swarm optimization (PSO) in order to solve FJSP. Different techniques of diversification have also been explored in order to improve the performance of our model. Our approach has been tested on a set of benchmarks existing in the literature. The results obtained show that the hybridization of TS and PSO led to promising results.
