EAI Endorsed Transactions on Industrial Networks and Intelligent Systems 15(4): e2

Research Article

Linking Data According to Their Degree of Representativeness (DoR)

Download68 downloads
  • @ARTICLE{10.4108/inis.2.4.e2,
        author={Fr\^{e}d\^{e}ric Blanchard and Amine A\~{n}t-Younes and Michel Herbin},
        title={Linking Data According to Their Degree of Representativeness (DoR)},
        journal={EAI Endorsed Transactions on Industrial Networks and Intelligent Systems},
        volume={15},
        number={4},
        publisher={ICST},
        journal_a={INIS},
        year={2015},
        month={6},
        keywords={Mining complex data, Representativeness, Graph based data analysis},
        doi={10.4108/inis.2.4.e2}
    }
    
  • Frédéric Blanchard
    Amine Aït-Younes
    Michel Herbin
    Year: 2015
    Linking Data According to Their Degree of Representativeness (DoR)
    INIS
    ICST
    DOI: 10.4108/inis.2.4.e2
Frédéric Blanchard1,*, Amine Aït-Younes1, Michel Herbin1
  • 1: Université de Reims Champagne-Ardenne, CReSTIC, UFR Sciences Exactes et Naturelles, Moulin de la Housse, BP 1039, 51687 Reims CEDEX 2, FRANCE
*Contact email: frederic.blanchard@univ-reims.fr

Abstract

This contribution addresses the problem of extracting some representative data from complex datasets and connecting them in a directed graph. First we define a degree of representativeness (DoR) inspired of the Borda voting procedure. Secondly we present a method to connect pairwise data using neighborhoods and the DoR as an objective function. We then present case studies as illustrative purposes: unsupervised grouping of binary images, analysis of co-authorships in a research team and structuration of a medical patient-oriented database.