Wilfried Probst

An agent-based approach for integrating user profile into a knowledge management process

This paper presents an approach based on a user agent to permit a number of users connected to distant machines to access different information sources in order to satisfy their requests. This user agent permits the simplification of the information search from distributed sources by making them transparent to the users. The agent considers the specific needs of each user during the search and responds with reference to their profile. It also permits the processing of one or more information requests by one or more users, as well as concurrent responses to each of them. Moreover, the agent provides its users with a measure of interaction, in order to enhance the quality and quantity of the results obtained. As a result, the agent is endowed with the ability to filter and refine the search, thus improving its service to the users.

An artificial neural network approach for routing in distributed computer networks

Abstract This paper presents a new approach based on the Hopfield model of artificial neural networks to solve the routing problem in a context of computer network design. The computer networks considered are packet switching networks, modeled as non-oriented graphs where nodes represent servers, hosts or switches, while bi-directional and symmetric arcs represent full duplex communication links. The proposed method is based on a network representation enabling to match each network configuration with a Hopfield neural network in order to find the best path between any node pair by minimizing an energy function. The results show that the time delay derived from flow assignment carried out by this approach is, in most cases, better than those determined using conventional routing heuristics. Therefore, this neural-network approach is suitable to be integrated into an overall topological design process of moderate-speed and high-speed networks subject to quality of service constraints as well as to changes in configuration and link costs.

Routing in computer networks using artificial neural networks

Abstract This paper proposes a heuristic approach based on Hopfield model of neural networks to solve the problem of routing which constitutes one of the key aspects of the topological design of computer networks. Adaptive to changes in link costs and network topology, the proposed approach relies on the utilization of an energy function which simulates the objective function used in network optimization while respecting the constraints imposed by the network designers. This function must converge toward a solution which, if not the best is at least as close as possible to the optimum. The simulation results reveal that the end-to-end delay computed according to this neural network approach is usually better than those determined by the conventional routing heuristics, in the sense that our routing algorithm realizes a better trade-off between end-to-end delay and running time, and consequently gives a better performance than many other well-known optimal algorithms.

Internet telephony signaling

Abstract The term “multimedia session” refers to the integration of data coming from various sources, such as sound, video and text, within a computer application. Telephony over the Internet is among the more exciting current developments. The signaling of a telephone call consists of the set of messages and procedures used to establish a connection, to request changes in communication bandwidth, to obtain the message status for the end points participating in the conversation, and to close the link. At present there exist two competing signaling protocols for Internet telephony, viz., the H.323 protocol sponsored by the ITU and the Session Invitation Protocol (SIP) sponsored by the IETF. Each of them supplies its own signaling mechanisms. In this paper, these two protocols in terms of their main functionalities are compared. Based on the results of this comparison, a Client/Server architecture for the development of an application that supports a basic SIP implementation, as well as the formulation of requests allowing the establishment and the disconnection of communications between a number of users in a multimedia session are then defined.

Modelling intelligent agents for information filtering

This paper presents a model of filtering agents for the search of information from heterogeneous sources. This model deals in particular with user requests and with source requests in the process of information filtering. Typically, a task of filtering is subdivided into sub-tasks, which are then executed by different agents; each endowed with a specific expertise. These agents are thus integrated into a multi-agent architecture. The communication between the agents is accomplished by the exchange of messages expressed in the KQML and ISAME languages, using previously-defined ontologies.

The computational object approach for network and systems management

Abstract In the past 20 years, Network and Systems Management (N&SM) has thrived on mostly centralized or weakly distributed paradigms. Advances in technologies and software engineering suggested new ways of doing N&SM. The computational object approach is one of the main management technologies that have recently appeared. Based on this approach, the architecture proposed in this paper uses programming concepts instead of protocol concepts and conceals protocol complexity by easily manipulated components. Concerning the information model, the Common Information Model (CIM) is used which is designed in an object-oriented manner that is the key behind scalable N&SM.The implementation is done in the Windows environment and using the C++ programming language.

An intelligent information-filtering assistant integrated into a multi-agent architecture

The great volume of information available on the Web and other sources, and the lack of tools to facilitate an intelligent selection of pertinent documents, make it difficult to extract the ones of interest to a user. We have therefore created and implemented an intelligent information-filtering assistant based on a multi-agent architecture, to support the user in the search for information. We have chosen a boolean search algorithm and our approach achieves an efficiency comparable to that of other information search engines.

Conceptual model for an agent utilised in information searching

For facilitating information searching from divided heterogeneous sources on a network, a modular approach was proposed, giving rise to a multi-agent system named ISAME. The system aims to offer to the agents representing the users, a simplified access to a dynamic collection of information sources available in electronic format. Thus the services aim to facilitate and optimise the process of information searching. The personalisation of the ISAME system for a user is effected by the intervention of the agents used. These agents realise a series of tasks and functions aimed at providing an information system regarding the different users. This article presents a conceptual model of agent user type from the objective of providing the maximum help and services to a class of users with ISAME. It results in a reduction of the server charge in the ISAME environment, before a certain number of operations and sub-tasks at the request of the users are effected locally.