Résumé: This talks presents to what extent the classification in terms of belief function and upper/lower probabilities carries over to qualitative fuzzy measures and possibility theory.

Résumé: Cet article présente des variantes de l’intégrale de Sugeno dans un cadre qualitatif fini. On introduit tout d’abord trois facons d’interpréter les poids d’importance associés aux groupes de critères, sur une algèbre de Heyting. On introduit ensuite l’idée de “désintégrale”, duale de celle d’intégrale. Lors de l’évaluation d’un objet, une désintégrale est maximale si aucun défaut eventuel n’est présenté de manière notable, tandis qu’une intégrale est maximale si tous les avantages eventuels sont suffisamment présents. Cette idée conduit à une représentation bipolaire des préférences, au moyen d’une paire constituée par une intégrale et une désintégrale.

Résumé: This paper presents several variants of Sugeno integral, and in particular the idea of (qualitative) desintegrals, a dual of integrals. When evaluating an item, desintegrals are maximal if no defects at all are present, while integrals are maximal if all advantages are sufficiently present. This idea leads to a bipolar representation of preferences, by means of a pair made of an integral and a desintegral, whose possibilistic logic counterparts are outlined (in the case where criteria are binary).

Résumé: Deconvolution consists of reconstructing a signal from blurred (and usuall noisy) sensory observations. It requires perfect knowledge of the impulse response of the sensor. Relevant works in the litterature propose methods with improved precision and robustness. But those methods are not able to account for a partial knowledge of the impulse response of the sensor. In this article, we experimentally show that inverting a Choquet capacity-based model of an imprecise knowledge of this impulse response allows to robustly recover the measured signal. The method we use is an interval valued extension of the well known Schultz procedure.

Ref HAL: hal-00783638_v1
Résumé: Dans cet article, nous portons notre regard sur l'aide à la décision du point de vue des systèmes décisionnels au sens des entrepôts de données et de l'analyse en ligne. Après avoir défini les concepts qui sous-tendent ces systèmes, nous nous proposons d'aborder les problématiques de recherche qui leur sont liées selon quatre points de vue : les données, les environnements de stockage, les utilisateurs et la sécurité. Nous abordons finalement les problèmes qui restent ouverts dans le domaine des entrepôts de données.

Résumé: In machine learning, the multiple classifiers aggregation problems consist in using multiple classifiers to enhance the quality of a single classifier. Simple classifiers as mean or majority rules are already used, but the aggregation methods used in voting theory or multi-criteria decision making should increase the quality of the obtained results. Meanwhile, these methods should lead to better interpretable results for a human decision-maker. We present here the results of a first experiment based on the use of Choquet integral, decisive sets and rough sets based methods on four different datasets.

---------

Résumé: Twitter have become a significant means by which people communicate with the world and describe their current activities, opinions and status in short text snippets. Tweets can be analyzed automatically in order to derive much potential information such as, interesting topics, social influence, user's communities, etc. Extraction communities within social networks has been a focus of recent work in several areas. Different from the most community discovery methods focused on the relations between users, we aim to derive user's communities based on common topics from user's tweets. For instance, if two users always talk about politic in their tweets, thus they can be grouped in the same community which is related to politic topic. To achieve this goal, we propose a new approach called CETD: Community Extraction based on Topic-Driven-Model. This approach combines our proposed model used to detect topics of the user's tweets based on a semantic taxonomy together with a community extraction method based on the hierarchical clustering technique. Our experimentation on the proposed approach shows the relevant of the users communities extracted based on their common topics and domains.

---------