Universität zu Köln

Rißler-Pipka, Nanette (2017). Image and Text in Numbers: Layout Analysis for Hispanic and Spanish Modern Magazines. In: Kodikologie und Paläographie im Digitalen Zeitalter 4 – Codicology and Palaeography in the Digital Age 4, pp. 25-42. Norderstedt: Books on Demand. ISBN 978-3-7448-3877-1

Preview

PDF 03_rissler-pipka.pdf - Published Version Download (14MB)

Abstract

Hispanic and Spanish modern magazines were long time a neglected field of research. Even if the magazines are regarded as a valuable source for information about contemporary cultural, social and political life for various disciplines like cultural, literary, media or social studies or linguistics. But stored in libraries as sensitive material and threated soon by decay most of the magazines were not accessible for researchers. Since digitization this has changed. The project Revistas culturales 2.0 (University of Augsburg) tries to work on the digital collection of the IAI (Ibero-American Institute, Preußischer Kulturbesitz, Berlin) by annotating the magazine pages and analysing the metadata using digital tools. But as we are experienced in cultural, media and romance studies, the complex field of automatic document analysis stayed hidden for us, without the cooperation with experts in computer science (where in context of digitizing projects and OCR important research is already achieved).1 Also, in the DH this field is not as advanced as for example text mining, therefore it proved difficult to find tools for quantitative analysis (for example relation of textual and image parts in the magazines). In cooperation with the project eCodicology and reusing their tool SWATI we found now a way how layout analysis for Hispanic and Spanish modern magazines might be done in future by measuring each page automatically. In consequence the paper can only present a concept of how useful the tool and quantitative analysis of the layout might be for analysing the Hispanic and Spanish modern magazines from the perspective of Humanities.