Altran

About: Altran is a company organization based out in Neuilly-sur-Seine, France. It is known for research contribution in the topics: Software development & Formal verification. The organization has 488 authors who have published 512 publications receiving 6395 citations.

Constrained Clustering Problems: NewOptimization Algorithms

Abstract: Constrained clustering problems are often considered in massive data clustering and analysis. They are used in modeling various issues in anomaly detection, classification, systems’ misbehaviour, etc. In this paper, we focus on generalizing the K-Means clustering approach when involving linear constraints on the clusters’ size. Indeed, to avoid local optimum clustering solutions which consists in empty clusters or clusters with few points, we propose linear integer programming approaches based on relaxation and rounding techniques to cope with scalability issues. We show the efficiency of the new proposed approach, and assess its performance using five data-sets from different domains.

Digital control room ergonomique

Abstract: ALTRAN a cherche a optimiser la communication entre les differents acteurs d'un projet complexe integrant differents membres en creant une interface graphique, intuitive et innovante. Dans une premiere etape, il a ete necessaire d'eliciter avec les utilisateurs finaux le niveau d'information exacte et necessaire a la conduite du projet. Dans une seconde etape, des experts en ergonomie, ont concu une interface homme machine pertinente et soutenant l'activite de supervision de projets complexes. Enfin, suite a un benchmark technologique, un outil a ete developpe et implante avec succes dans le management de la performance de programme aeronautique.

A Self Organizing OneM2M IoT Network

Abstract: In IoT, there is a huge diversity in types of devices, protocols and mechanisms that must be supported. ‘IoTification’ helps these objects to communicate with each other, exchange information in order to have a better connected and managed system. In order to reduce the number of errors during manual maintenance interventions on the IoT network infrastructure and to improve the efficiency of reconfiguration actions by automatic mechanisms, a self-organizing network (SON)-based architecture is proposed for a oneM2M IoT network. This makes it possible to take into account the heterogeneity of IoT subnets and to reduce the costs related to network infrastructure extensions. The proposed approach is implemented and evaluated according to the ACS quality management system.

Unabhängigkeits- und Homogenitätstests

Abstract: In diesem Kapitel wird der \( {\textit{X}}^2\)-Test und der auf R.A. Fisher zuruckgehende exakte Test auf Unabhangigkeit von Merkmalen behandelt. Der \( {\textit{X}}^2\)-Unabhangigkeitstest basiert direkt auf dem \( {\textit{X}}^2\)-Anpassungstest, und die entsprechenden p-Werte werden bei Vorliegen „groserer“ Fallzahlen naherungsweise berechnet. Dagegen liefert „Fishers exakter Test“, der vorzugsweise bei kleineren Trefferzahlen angewendet wird, eine genaue Berechnung der fraglichen Wahrscheinlichkeiten. In Homogenitatstests wird die Verteilungshomogenitat bezuglich mehrerer Stichproben untersucht. Auch wenn solche Tests grundsatzlich von Unabhangigkeitstests zu unterscheiden sind, werden wir Versionen kennenlernen, die rechnerisch vollig analog zum \( {\textit{X}}^2\)-Test und zu Fishers exaktem Test verlaufen.