piwik-script

Intern
    DMIR Research Group

    Publications by Andreas Hotho

    These publications are hosted by BibSonomy.

    Data Mining on Folksonomies

    in Intelligent Information Access . Armano, Giuliano; de Gemmis, Marco; Semeraro, Giovanni; Vargiu, Eloisa ( Hrsg. ), Volume   301   von   Studies in Computational Intelligence 301 , Springer , Berlin / Heidelberg , 2010 . Seite 57-82 .

    Social resource sharing systems are central elements of the Web 2.0 and use all the same kind of lightweight knowledge representation, called folksonomy. As these systems are easy to use, they attract huge masses of users. Data Mining provides methods to analyze data and to learn models which can be used to support users. The application and adaptation of known data mining algorithms to folksonomies with the goal to support the users of such systems and to extract valuable information with a special focus on the Semantic Web is the main target of this paper. In this work we give a short introduction into folksonomies with a focus on our own system BibSonomy. Based on the analysis we made on a large folksonomy dataset, we present the application of data mining algorithms on three different tasks, namely spam detection, ranking and recommendation. To bridge the gap between folksonomies and the Semantic Web, we apply association rule mining to extract relations and present a deeper analysis of statistical measures which can be used to extract tag relations. This approach is complemented by presenting two approaches to extract conceptualizations from folksonomies.
    Weitere Informationen
    Herausgeber Armano, Giuliano; de Gemmis, Marco; Semeraro, Giovanni; Vargiu, Eloisa
    DOIhttp://dx.doi.org/10.1007/978-3-642-14000-6_4
    Tags2010  data  folksonomies  for:dmir  mining  myown 

    Hinweis zum Datenschutz

    Mit 'OK' verlassen Sie die Seiten der Universität Würzburg und werden zu Facebook weitergeleitet. Informationen zu den dort erfassten Daten und deren Verarbeitung finden Sie in deren Datenschutzerklärung.

    Hinweis zum Datenschutz

    Mit 'OK' verlassen Sie die Seiten der Universität Würzburg und werden zu Twitter weitergeleitet. Informationen zu den dort erfassten Daten und deren Verarbeitung finden Sie in deren Datenschutzerklärung.

    Social Media
    Kontakt

    Andreas Hotho
    DMIR Research Group
    Am Hubland
    97074 Würzburg

    Tel.: +49 931 31-86731
    Fax: +49 931 31-86732

    Suche Ansprechpartner

    Hubland Süd, Geb. M2
    Hubland Süd, Geb. M2