Conformément aux principes de Linked Data, les fournisseurs de données ont publié des milliards de faits en tant que données RDF. Exécuter les requêtes SPARQL sur les endpoints SPARQL ou les serveurs Triple Pattern Fragments (TPF) permet de consommer facilement des données du Linked Data. Cependant, le traitement des requêtes SPARQL fédérées, tout comme le traitement des requêtes TPF, décompose la requête initiale en de nombreuses sous-requêtes. Les fournisseurs de données ne voient alors que les sous-requêtes et la requête initiale n’est connue que des utilisateurs finaux. La connaissance des requêtes exécutées est fondamentale pour les fournisseurs, afin d’assurer un contrôle de l’utilisation des données, d’optimiser le coût des réponses aux requêtes, de justifier un retour sur investissements, d’améliorer l’expérience utilisateur ou de créer des modèles commerciaux à partir de tendances d’utilisation. Dans cette thèse, nous nous concentrons sur l’analyse des logs d’exécution des serveurs TPF et des endpoints SPARQL pour extraire les Basic Graph Patterns (BGP) des requêtes SPARQL exécutées. Le principal défi pour l’extraction des BGPs est l’exécution simultanée des requêtes SPARQL. Nous proposons deux algorithmes : LIFT et FETA. Sous certaines conditions, nous constatons que LIFT et FETA sont capables d’extraire des BGPs avec une bonne précision et un bon rappel. / Following the principles of Linked Data, data providers published billions of facts as RDF data. Executing SPARQL queries over SPARQL endpoints or Triple Pattern Fragments (TPF) servers allow to easily consume Linked Data. However, federated SPARQL query processing and TPF query processing decompose the initial query into subqueries. Consequently, the data providers only see subqueries and the initial query is only known by end users. Knowing executed SPARQL queries is fundamental for data providers, to ensure usage control, to optimize costs of query answering, to justify return of investment, to improve the user experience or to create business models of usage trends. In this thesis, we focus on analyzing execution logs of TPF servers and SPARQL endpoints to extract Basic Graph Patterns (BGP) of executed SPARQL queries. The main challenge to extract BGPs is the concurrent execution of SPARQL queries. We propose two algorithms: LIFT and FETA. LIFT extracts BGPs of executed queries from a single TPF server log. FETA extracts BGPs of federated queries from a log of a set of SPARQL endpoints. For experiments, we run LIFT and FETA on synthetic logs and real logs. LIFT and FETA are able to extract BGPs with good precision and recall under certain conditions.

Le contexte économique actuel fait naître de nouveaux besoins chez les entreprises. L’élargissement du marché et la multiplication de la concurrence impliquent de nouveaux objectifs pour les industriels et doivent adapter leur mode de fonctionnement et leur organisation pour les atteindre. Les efforts se portent sur la minimisation de la distance existante entre concepteurs et clients. Cet éloignement est souvent à l’origine de mauvaises compréhensions et de dérivations par rapport à la demande client initiale. Désormais, le client est placé au cœur du processus de conception pour mieux intégrer ses exigences et ainsi satisfaire ses besoins. Ces nouvelles pratiques se démocratisent grâce à l’évolution du web vers le web 2.0 facilitant la communication et la formation de réseaux sociaux. Les systèmes de management de l’information actuellement implémentés dans les entreprises présentent des limites. Les systèmes PLM gérant les informations relatives au produit tout au long de son cycle de vie, traitent peu de l’interaction utilisateur/produit qui pourtant s’avère porteuse d’informations stratégiques pour l’amélioration de la conception et l’innovation de produits. En considérant l’interaction entre un utilisateur et un produit dans un contexte donné, il est possible d’identifier les contraintes liées aux usages de cet utilisateur. C’est sur ces informations d’usage que notre recherche s’est orienté afin d’améliorer la compréhension des besoins clients et déceler des idées innovantes de produits. En élargissant la notion d’utilisateur plutôt associé au client, nous souhaitons y inclure également les employés, les revendeurs et détaillants qui sont à leurs tours à un moment donnée du cycle de vie du produit des utilisateurs. Dans cette optique, nous proposons une approche appelée ULM (Usage Lifecycle Management) qui s’inscrit dans une dynamique de management des usages de divers produits, utilisateurs et environnements. Notre travail s’est axé sur la phase de capitalisation de ces usages à travers les processus aval du cycle de vie, utilisation/commercialisation, maintenance et recyclage. Pour cela, nous proposons l’approche ULM au travers d’une plateforme web externalisée qui se situe au carrefour de plusieurs utilisateurs. Cet outil baptisé C 3 pour Conception Consommation Communication s’oriente autour d’un trièdre utilisateur/produit/contexte pour capitaliser et formaliser les informations d’usage d’un produit. C’est en pensant conjointement les usages et les concepts qu’il sera possible d’aboutir à des produits innovants capables de faire la différence sur un marché ultra concurrentiel. / The current economical context implies news needs for companies. Market spread and competitor’s multiplication imply new objectives for companies which must adapt their functioning process and organization to succeed. Efforts focus on minimizing the distance between designers and customers. This distance often originates misunderstandings and deviations related to the initial customer’s demands. The customer is now placed at the core of the design process in order to better integrate his expectations and satisfy his needs. These new practices democratize themselves thanks to the web evolution towards web 2.0, which facilitates communication and triggered the apparition of social networks. Current information management system implement in companies have limitations. Product-Life Management (PLM) systems manage product-related information over its lifetime. They only slightly encompass the user/product interactions, hence missing strategic information for improving products design and innovations. We consider the interactions between a user and a product within a given context and identify the constraints linked to the use of the product. We oriented our research toward this usage information in order to detect the customer’s needs and better initiate product-related innovations. We widen the notion of user to include employees, retailers and suppliers who punctually interact with the product in its lifecycle. To address this issue, we suggest an approach named ULM (Usage Lifecycle Management). This approach involves a new dynamic for managing usage for several product, users and environments. Our work is axed on the usages capitalization phase through downstream lifecycle processes, use/commercialization, maintenance and recycling. We hence concretize our ULM approach trough an outsourced web platform placed as a center point for users. This tool, named C³ for Conception Consumption Communication, is articulated around the user/product/context triangle to capitalize and formalize product usage information. Only through joint reflexion over usages and concepts will it be to reach innovative products able to make a difference in an ultra-competitive market.

No description available.

No description available.

The purpose of this study was to identify some environmental and overt behavioral factors which are characteristic of drug users. This was attempted to enable individuals who work with adolescents to identify potential users.

MSc (Med), Epidemiology and Biostatistics, Faculty of Health Sciences, University of the Witwatersrand / Background: The knowledge of contraceptive use is high among men and women in South Africa. However, contraceptive prevalence rate is moderate and unplanned pregnancies are common. Understanding the determinants of contraceptive use and unplanned pregnancy will inform future interventions that aim to maintain consistent contraceptive use and reduce unplanned pregnancies. Aim: The study aims to describe factors associated with contraceptive use and unplanned pregnancy in the South African population. Methods: A secondary data analysis was carried out on data collected in a cross–sectional survey conducted in Potchefstroom, South Africa between August 2007 and March 2008.Results: Contraceptive prevalence was 69.5% and unplanned pregnancy was 59.7%. The risk factors for contraceptive use included woman’s employment status at the last pregnancy, woman’s partner employment status at the last pregnancy and number of miscarriages a woman had experienced. The risk factors for unplanned pregnancy included race, woman’s age , education level and employment status at last pregnancy, number of miscarriages, contraceptive use and partner’s employment status at last pregnancy. Conclusion: If the prevalence of unplanned pregnancies is to be reduced, policies and programmes need to address economic factors which were associated with both contraceptive use and unplanned pregnancy. Further study needs to be carried out as to the reasons behind why a woman with a previous history of a miscarriage is less likely to have an unplanned pregnancy yet she is less likely to be on contraception.

Estágio realizado na AuditMark e orientado pelo Eng.º Pedro Fortuna / Tese de mestrado integrado. Engenharia Informática e Computação. Faculdade de Engenharia. Universidade do Porto. 2010

The purpose of this article is to gain insight in the South African tobacco industry, its influence on the community and the role of anti-tobacco campaigns. This is done by conducting an empirical investigation of the attitudes and behaviour of smokers and non-smokers regarding the national government campaigns against tobacco products. Consequently, it will contribute in determining the efficiency of the anti-tobacco campaigns as well as the effect of the Tobacco Products Control Amendment Act, 1999 (Act 12 of 1999) on smokers and non-smokers. It is postulated that by applying the principles of marketing (demarketing and remarketing) it could contribute significantly in influencing smoking behaviour amongst South Africans.

Th.--Méd.--Reims, 1980. N°: 24.

No description available.