The increasing amounts of electronic data of all forms, produced by humans (e.g. Web pages, structured content such as Wikipedia or the blogosphere etc.) and/or automatic tools (loggers, sensors, Web services, scientific programs or analysis tools etc.) leads to a situation of unprecedented potential for extracting new knowledge, finding new correlations, or simply making sense of the data.Visual analytics aims at combining interactive data visualization with data analysis tasks. Given the explosion in volume and complexity of scientific data, e.g., associated to biological or physical processes or social networks, visual analytics is called to play an important role in scientific data management.Most visual analytics platforms, however, are memory-based, and are therefore limited in the volume of data handled. Moreover, the integration of each new algorithm (e.g. for clustering) requires integrating it by hand into the platform. Finally, they lack the capability to define and deploy well-structured processes where users with different roles interact in a coordinated way sharing the same data and possibly the same visualizations.This work is at the convergence of three research areas: information visualization, database query processing and optimization, and workflow modeling. It provides two main contributions: (i) We propose a generic architecture for deploying a visual analytics platform on top of a database management system (DBMS) (ii) We show how to propagate data changes to the DBMS and visualizations, through the workflow process. Our approach has been implemented in a prototype called EdiFlow, and validated through several applications. It clearly demonstrates that visual analytics applications can benefit from robust storage and automatic process deployment provided by the DBMS while obtaining good performance and thus it provides scalability.Conversely, it could also be integrated into a data-intensive scientific workflow platform in order to increase its visualization features.

L'expansion du World Wide Web et la multiplication des sources de données (capteurs, services Web, programmes scientifiques, outils d'analyse, etc.) ont conduit à la prolifération de données hétérogènes et complexes. La phase d'extraction de connaissance et de recherche de corrélation devient ainsi de plus en plus difficile.Typiquement, une telle analyse est effectuée en utilisant les outils logiciels qui combinent: des techniques de visualisation, permettant aux utilisateurs d'avoir une meilleure compréhension des données, et des programmes d'analyse qui effectuent des opérations d'analyses complexes et longues.La visualisation analytique (visual analytics) vise à combiner la visualisation des donnéesavec des tâches d'analyse et de fouille. Etant donnée la complexité et la volumétrie importante des données scientifiques (par exemple, les données associées à des processus biologiques ou physiques, données des réseaux sociaux, etc.), la visualisation analytique est appelée à jouer un rôle important dans la gestion des données scientifiques.La plupart des plateformes de visualisation analytique actuelles utilisent des mécanismes en mémoire centrale pour le stockage et le traitement des données, ce qui limite le volume de données traitées. En outre, l'intégration de nouveaux algorithmes dans le processus de traitement nécessite du code d'intégration ad-hoc. Enfin, les plate-formes de visualisation actuelles ne permettent pas de définir et de déployer des processus structurés, où les utilisateurs partagent les données et, éventuellement, les visualisations.Ce travail, à la confluence des domaines de la visualisation analytique interactive et des bases de données, apporte deux contributions. (i) Nous proposons une architecture générique pour déployer une plate-forme de visualisation analytique au-dessus d'un système de gestion de bases de données (SGBD). (ii) Nous montrons comment propager les changements des données dans le SGBD, au travers des processus et des visualisations qui en font partie. Notre approche permet à l'application de visualisation analytique de profiter du stockage robuste et du déploiement automatique de processus à partir d'une spécification déclarative, supportés par le SGBD.Notre approche a été implantée dans un prototype appelé EdiFlow, et validée à travers plusieurs applications. Elle pourrait aussi s'intégrer dans une plate-forme de workflow scientifique à usage intensif de données, afin d'en augmenter les fonctionnalités de visualisation. / The increasing amounts of electronic data of all forms, produced by humans (e.g. Web pages, structured content such as Wikipedia or the blogosphere etc.) and/or automatic tools (loggers, sensors, Web services, scientific programs or analysis tools etc.) leads to a situation of unprecedented potential for extracting new knowledge, finding new correlations, or simply making sense of the data.Visual analytics aims at combining interactive data visualization with data analysis tasks. Given the explosion in volume and complexity of scientific data, e.g., associated to biological or physical processes or social networks, visual analytics is called to play an important role in scientific data management.Most visual analytics platforms, however, are memory-based, and are therefore limited in the volume of data handled. Moreover, the integration of each new algorithm (e.g. for clustering) requires integrating it by hand into the platform. Finally, they lack the capability to define and deploy well-structured processes where users with different roles interact in a coordinated way sharing the same data and possibly the same visualizations.This work is at the convergence of three research areas: information visualization, database query processing and optimization, and workflow modeling. It provides two main contributions: (i) We propose a generic architecture for deploying a visual analytics platform on top of a database management system (DBMS) (ii) We show how to propagate data changes to the DBMS and visualizations, through the workflow process. Our approach has been implemented in a prototype called EdiFlow, and validated through several applications. It clearly demonstrates that visual analytics applications can benefit from robust storage and automatic process deployment provided by the DBMS while obtaining good performance and thus it provides scalability.Conversely, it could also be integrated into a data-intensive scientific workflow platform in order to increase its visualization features.

Doctoral thesis deals with parameters of LED light sources and their static and dynamic changes due to variable operational conditions. Firstly there is briefly described the history of light emitting diodes, that brought LED from unintentional observation of luminescence to light source, which is generally used through the whole area of lighting techniques. Thesis focuses on light parameters analysis, which are specific for semiconductor light sources and their mutual relations. These relations in compliance with understanding the principles of light emission in semiconductor light sources enable establishing of basic connections and effects usable for LED behaviour model. Mutual relations of LED parameters and influencing factors are verified by measurement of chosen testing samples. For deeper understanding of diode parameter influencing mechanisms there are analysed current degradation models including degradation of partial components. On the basis of these facts there is created a model of LED luminaire operating with LED parameters together with cooling system that represent the integral part of the luminaire. Impact of supply drivers is also studied, because they are necessary for LED operation. Designed model supports the area of static changes of light parameters, as well as the dynamic changes of output parameters. Presented model is verified by testing measurements at basic operational states and partially outside of the common operational limits. There are also presented the possibilities of real use of the model that involve the evaluation of designed luminaire for specific light sources, searching for optimal operational limits, optimization of luminaire heatsink or using public lighting systems for regulation of electricity grid load fluctuations.