Abstract

With the increasing complexity of Web service management in highly dynamic environments, service providers must participate to the self-adaptation process as they are aware of the capabilities and requirements of their executing services. However, defining additional adaptation policies demands a greater effort. A primary condition to reuse other providers’ policies is to understand the information specifying a service behavior. Despite several efforts, a common self-management syntax and semantic was not achieved. In this paper, we propose an ontology-based categorization of autonomic service metadata, which will be used not only for policy discovery and integration, but also for the composition of self-adaptive Web services. We, then, propose to semantically annotate providers’ policies. This allows providers to discover and reuse policies based on their semantic meaning. Finally, we propose a framework for the discovery, matching, and integration of providers’ policies.

Abstract

In line with the multi-disciplinary growing interest in emotions and the scientific proof of their usefulness for taking decisions, scholars, in agent-oriented systems, start to account for emotions when building upon intelligence and realism in rational agents. As a result, several computational models of emotions were developed and new architectures for emotional artificial agents were proposed, in particular the Emotional Belief Desire Intention (EBDI) agents. In this paper, we provide a comprehensive description of two computational models which are used to generate immediate and expected emotions. These models will be incorporated within an EBDI agent architecture that takes into consideration these two types of emotions.

Abstract

Recommending personalized events from the huge amount of information on the social web is a challenging problem. Besides, dealing with such recommendation improves significantly the professional communities activities. Likewise, it helps them to make appropriate decisions while saving time and efforts. In this paper, we propose a hybrid recommender system which suggests suitable events to the HanDicraft Women (HDWs) from Tunisia and Algeria. The established system considers the final user needs and demographic attributes. Indeed, it combines the knowledge-based and demographic approaches together. Useful information related to the HDWs and available events are represented through a semantic formalization : Friend Of A Friend (FOAF) Ontoloy and Online presence Ontology (OPO). Moreover, the recommender is able to manage the dynamicity and heterogeneity of the HDWs environments since it is based on a multi-agent architecture.

Abstract

The online supply requirement management within a heterogonous environment represents a real challenge. While the traditional e-purchase is widely adopted, the e-barter seems to be an ambitious alternative. It is mainly solicited when suppliers might be unavailable or the delivery timeouts are important. Moreover, it reinforces the communication between the producer and his professional network. In this paper, we propose an emulation of the handicraft women e-procurement process based on the power of multi-agent paradigm and ontology formalism. Indeed, we establish an e-barter framework which targets to recommend in real time, under different circumstances and regarding the handicraft woman situation the suitable exchange partners. Likewise, we established several producing rules in order to deduce automatically the best sourcing moment. Furthermore, the handicraft woman which is the decision maker might drive an e-barter auction (e-BA) process in order to choose the best exchange opportunities and then minimize her expenses. We consider the e-BA as a new concept merging the barter and auction notions.

Abstract

Abstract

Inspired by the idea of interacting intelligent agents of a multi-agent system, we introduce a multi-agent based optimization method applied to the quadratic assignment problem (MAOM-QAP). MAOM-QAP is composed of several agents (decision-maker agent, local search agents, crossover agents and perturbation agent) which are designed for the purpose of intensified and diversified search activities. With the help of a reinforcement learning mechanism, MAOM-QAP dynamically decides the most suitable agent to activate according to the state of search process. Under the coordination of the decision-maker agent, the other agents fulfill dedicated search tasks. The performance of the proposed approach is assessed on the set of well-known QAP benchmark instances, and compared with the most advanced QAP methods of the literature. The ideas proposed in this work are rather general and could be adapted to other optimization tasks. This work opens the way for designing new distributed intelligent systems for tackling other complex search problems.

Abstract

Detecting changes in data streams attracts major attention in cognitive computing systems. The challenging issue is how to monitor and detect these changes in order to preserve the model performance during complex drifts. By complex drift, we mean a drift that presents many characteristics in the sometime. The most challenging complex drifts are gradual continuous drifts, where changes are only noticed during a long time period. Moreover, these gradual drifts may also be local, in the sense that they may affect a little amount of data, and thus make the drift detection more complicated. For this purpose, a new drift detection mechanism, EDIST2, is proposed in order to deal with these complex drifts. EDIST2 monitors the learner performance through a self-adaptive window that is autonomously adjusted through a statistical hypothesis test. This statistical test provides theoretical guarantees, regarding the false alarm rate, which were experimentally confirmed. EDIST2 has been tested through six synthetic datasets presenting different kinds of complex drift, and five real-world datasets. Encouraging results were found, comparing to similar approaches, where EDIST2 has achieved good accuracy rate in synthetic and real-world datasets and has achieved minimum delay of detection and false alarm rate.

Abstract

Detecting changes in data streams attracts major attention in cognitive computing systems. The challenging issue is how to monitor and detect these changes in order to preserve the model performance during complex drifts. By complex drift, we mean a drift that presents many characteristics in the sometime. The most challenging complex drifts are gradual continuous drifts, where changes are only noticed during a long time period. Moreover, these gradual drifts may also be local, in the sense that they may affect a little amount of data, and thus make the drift detection more complicated. For this purpose, a new drift detection mechanism, EDIST2, is proposed in order to deal with these complex drifts. EDIST2 monitors the learner performance through a self-adaptive window that is autonomously adjusted through a statistical hypothesis test. This statistical test provides theoretical guarantees, regarding the false alarm rate, which were experimentally confirmed. EDIST2 has been tested through six synthetic datasets presenting different kinds of complex drift, and five real-world datasets. Encouraging results were found, comparing to similar approaches, where EDIST2 has achieved good accuracy rate in synthetic and real-world datasets and has achieved minimum delay of detection and false alarm rate.

Abstract

Because of the rapid growth of air traffic, optimizing airport management is becoming necessary in order to improveairport’s capacity and better align its resources to the received traffic. In this paper we study the assignment of the arriving aircrafts to the available gates using the fixed daily schedule. We introduce a new approach based on Genetic Algorithms (GA) to solve the gate assignment problem (GAP). The encoding strategy consists in representing the chromosome by a vector of integers. The index of each gene represents the flight number and its value represents the gate to which the flight will be assigned. The method used to generate the initial population is based on three different heuristics and a random sorting of the gates. The selection method is the “In fitness proportionate selection” known as “roulette wheel selection”. In addition to one point and two point Crossover operators, we designed a Greedy procedure Crossover (GPX) operator. The experimentation is based on the use of fictive scenarios generated in accordance with the physical characteristics of the Tunis Carthage Airport and using different flight schedules. The comparison between deterministic approach, simple heuristics and the GA has shown the efficiency of the last approach in terms of solution’s quality when we aim at solving the problems of large size. In order to determine the best configuration of the GA, we compared the different crossover operators and we noticed that the use of GPX improves the speed of convergence of the algorithm towards better solutions.