Direkt zur Hauptnavigation springen Direkt zum Inhalt springen
info(at)avldigital.remove-this.de
Deutsch English
Letzte Suchanfragen

Ergebnisse für *

Zeige Ergebnisse 1 bis 1 von 1.

  1. No Ground Truth?
    No Problem : Improving Administrative Data Linking Using Active Learning and a Little Bit of Guile
    Autor*in: Tahamont, Sarah; Jelveh, Zubin; McNeill, Melissa; Yan, Shi; Chalfin, Aaron; Hansen, Benjamin
    Erschienen: April 2023
    Verlag:  National Bureau of Economic Research, Cambridge, Mass

    While linking records across large administrative datasets ["big data"] has the potential to revolutionize empirical social science research, many administrative data files do not have common identifiers and are thus not designed to be linked to... mehr

    Zugang:
    Verlag (lizenzpflichtig)
    Resolving-System (lizenzpflichtig)

    Standort:
    Sächsische Landesbibliothek - Staats- und Universitätsbibliothek Dresden
    Fernleihe:
    keine Fernleihe

    Standort:
    Universitätsbibliothek Freiburg
    Fernleihe:
    keine Fernleihe

    Standort:
    Helmut-Schmidt-Universität, Universität der Bundeswehr Hamburg, Universitätsbibliothek
    Fernleihe:
    keine Fernleihe

    Standort:
    Staats- und Universitätsbibliothek Hamburg Carl von Ossietzky
    Fernleihe:
    keine Fernleihe

    Standort:
    Technische Informationsbibliothek (TIB) / Leibniz-Informationszentrum Technik und Naturwissenschaften und Universitätsbibliothek
    Fernleihe:
    keine Fernleihe

    Standort:
    ZBW - Leibniz-Informationszentrum Wirtschaft, Standort Kiel
    Fernleihe:
    keine Fernleihe

     

    While linking records across large administrative datasets ["big data"] has the potential to revolutionize empirical social science research, many administrative data files do not have common identifiers and are thus not designed to be linked to others. To address this problem, researchers have developed probabilistic record linkage algorithms which use statistical patterns in identifying characteristics to perform linking tasks. Naturally, the accuracy of a candidate linking algorithm can be substantially improved when an algorithm has access to "ground-truth" examples -- matches which can be validated using institutional knowledge or auxiliary data. Unfortunately, the cost of obtaining these examples is typically high, often requiring a researcher to manually review pairs of records in order to make an informed judgement about whether they are a match. When a pool of ground-truth information is unavailable, researchers can use "active learning" algorithms for linking, which ask the user to provide ground-truth information for select candidate pairs. In this paper, we investigate the value of providing ground-truth examples via active learning for linking performance. We confirm popular intuition that data linking can be dramatically improved with the availability of ground truth examples. But critically, in many real-world applications, only a relatively small number of tactically-selected ground-truth examples are needed to obtain most of the achievable gains. With a modest investment in ground truth, researchers can approximate the performance of a supervised learning algorithm that has access to a large database of ground truth examples using a readily available off-the-shelf tool

     
    Export in Literaturverwaltung   RIS-Format
      BibTeX-Format
    Quelle: Verbundkataloge
    Sprache: Englisch
    Medientyp: Buch (Monographie)
    Format: Online
    Schriftenreihe: NBER working paper series ; no. w31100
    Schlagworte: Statistische Daten; Big Data; Metadaten; Statistische Methode; Statistical Simulation Methods: General; Other Computer Software
    Umfang: 1 Online-Ressource, illustrations (black and white)
    Bemerkung(en):

    Hardcopy version available to institutional subscribers