Abstract Due to limited resources of individual senor nodes, there is usually a trade-off between energy spending forpackets transmissions and the appropriate level of reliability. Since link failures and packet losses are unavoidable, sensornetworks may tolerate a certain level of reliability without significantly affecting packets delivery performance and dataaggregation accuracy in favor of efficient energy consumption. However a certain degree of... Due to limited resources of individual senor nodes, there is usually a trade-off between energy spending forpackets transmissions and the appropriate level of reliability. Since link failures and packet losses are unavoidable, sensornetworks may tolerate a certain level of reliability without significantly affecting packets delivery performance and dataaggregation accuracy in favor of efficient energy consumption. However a certain degree of reliability is needed, especiallywhen hop count increases between source sensor nodes and the base station as a single lost packet may result in loss of alarge amount of aggregated data along longer hops. An effective solution is to trade off between energy and reliability whileimproving packet delivery and minimizing unnecessary packet transmissions in favor of high success reception rates ofrepresentative data packets. Based on this approach, the proposed scheme achieves moderate energy consumption and highpacket delivery ratio. The proposed scheme was experimentally investigated on a testbed of TelosB motes and proven to bemore robust and energy efficient than the current implementation of TinyOS2.x MultihopLQI. Read More Read Less

Abstract Over the past decade, mobile has experienced a revolution that will ultimately change the way we communicate.All these technologies have a common denominator exploitation of computer information systems, but their operation can betedious because of problems with heterogeneous data sources.To overcome the problems of heterogeneous data sources, we propose to use a technique of adding an extra layer interfacingapplications of... Over the past decade, mobile has experienced a revolution that will ultimately change the way we communicate.All these technologies have a common denominator exploitation of computer information systems, but their operation can betedious because of problems with heterogeneous data sources.To overcome the problems of heterogeneous data sources, we propose to use a technique of adding an extra layer interfacingapplications of management or supervision at the different data sources. This layer will be materialized by the implementationof a mediator between different host applications and information systems frequently used hierarchical and relational mannersuch that the heterogeneity is completely transparent to the VoIP platform. Read More Read Less

Abstract This paper presents an approach toward reducing multidimensional data in a data warehouse and giving theanalyst more facilities when querying data cubes. This reduction concerns the number of dimensions as well as the hugeamount of facts in the data warehouse. To make it possible, this approach combines the potential of the Principal ComponentAnalysis which is a purely statistical approach with... This paper presents an approach toward reducing multidimensional data in a data warehouse and giving theanalyst more facilities when querying data cubes. This reduction concerns the number of dimensions as well as the hugeamount of facts in the data warehouse. To make it possible, this approach combines the potential of the Principal ComponentAnalysis which is a purely statistical approach with the Rough Set Theory which is an Artificial Intelligence one. In fact, ThePrincipal Component Analysis is used, here, to reduce dimensions in the data warehouse without loosing information. So itcomputes a subset of the most important and pertinent dimensions. This subset of data dimensions is used by the Rough SetTheory in order to delete superfluous and redundant facts. The interest of this approach is to help user to focus on the mostimportant information of the data warehouse which is a very hard task when working on the whole multidimensionaldatabase. Read More Read Less

Abstract As a young discipline at the junction of computer science, artificial intelligence and cognitive sciences, knowl-edge engineering aims at modelling knowledge of a specific domain to operationalise them in a computer system. To this end,it offers theoretical tools, models and empirical methodologies to support Knowledge sharing between the user and thesystem. The work developed here is related to knowledge engineering... As a young discipline at the junction of computer science, artificial intelligence and cognitive sciences, knowl-edge engineering aims at modelling knowledge of a specific domain to operationalise them in a computer system. To this end,it offers theoretical tools, models and empirical methodologies to support Knowledge sharing between the user and thesystem. The work developed here is related to knowledge engineering of a particular type of system: case-based reasoningsystems (CBR) The case-based reasoning (CBR) is to solve a problem by remembering and adapting past cases alreadyresolved. The CBR systems handle various kinds of knowledge: the case, the domain knowledge, knowledge of similarity andadaptation. The cases are collected in a gradual manner when using the system and the case base is enriched incrementally,while other types of knowledge are typically acquired when the system design. In particular the domain knowledge.This paper presents an approach for acquiring domain knowledge-based adaptation system failures. This approach has beenimplemented in a prototype, called FRAKAS, using the description logic. Read More Read Less

Abstract In recent years, ontologies have played a key technology role for information sharing and agents interoperabilityin different information systems. But, it seems that there is always more than one conceptualization for the same domain oreven for similar domains. In other words, it emerges every day, new different ontology to model the same domain. Therefore,to answer queries on the modeled domain,... In recent years, ontologies have played a key technology role for information sharing and agents interoperabilityin different information systems. But, it seems that there is always more than one conceptualization for the same domain oreven for similar domains. In other words, it emerges every day, new different ontology to model the same domain. Therefore,to answer queries on the modeled domain, bridge the gaps between different ontologies is a key challenge for the researchersin the AI community by using ontology merging. In this paper, we propose an architecture for a semi-automatic ontologymerging process. The semi-automatic character is handled by the human intervention where the knowledge engineer inter-venes to validate the results provided by the similarity computation module. This later is based on a lexicosemantic algo-rithm that combines lexical and semantic measures to identify the similar concepts that have to be merged into a single onein the merged ontology, after human validation. The judged different concepts are directly copied to the merged ontology. Read More Read Less