Akademska digitalna zbirka SLovenije - logo
VSE knjižnice (vzajemna bibliografsko-kataložna baza podatkov COBIB.SI)
  • Document categorization based on OCR technology : an overview
    Zelenika, Darko ; Povh, Janez, 1973- ; Dobrovoljc, Andrej, 1967-
    In this paper a brief overview of document categorization process is presented with the focus on documents obtained by OCR (Optical Character Recognition) technology. Work of different authors from ... area of document categorization is described. Text obtained by OCR needs to be prepared in a way that categorization algorithms can use it to provide better categorization accuracy, thus such methods are introduced. A comparison of results of different categorization algorithms is shown. Most authors obtained the best results with SVM (Support Vector Machine) classifier. Two document categorization software programs are introduced both commercial and open source. An invoice recognition project on which authors of this paper are working on is introduced.
    Vir: Recent advances in information science : proceedings of the 7th European Computing Conference (EEC '13), Dubrovnik, Croatia, June 25-27, 2013 (Str. 409-414)
    Vrsta gradiva - prispevek na konferenci ; neleposlovje za odrasle
    Leto - 2013
    Jezik - angleški
    COBISS.SI-ID - 2048217363

Vnos na polico