05 Fakultät Informatik, Elektrotechnik und Informationstechnik

Permanent URI for this collectionhttps://elib.uni-stuttgart.de/handle/11682/6

Browse

Filters

2017 - 201952020 - 20231

Settings

Institute: search.filters.UBSInstitut.Institut für LiteraturwissenschaftSubject: search.filters.subject.004

Search Results

Now showing 1 - 6 of 6
  • Thumbnail Image
    ItemOpen Access
  • No Thumbnail Available
    ItemOpen Access
    CUTE - CRETA Un-/Shared Task zu Entitätenreferenzen
    (2017) Reiter, Nils; Blessing, André; Echelmeyer, Nora; Kremer, Gerhard; Koch, Steffen; Murr, Sandra; Overbeck, Maximilian; Pichler, Axel
    Dies ist die Veröffentlichung eines shared/unshared Task Workshops (entwickelt in CRETA: Center for Reflected Text Analytics), der im Rahmen der DHd 2017 in Bern (CH) stattfand. Im Gegensatz zu shared tasks, bei denen die Performanz verschiedener Systeme/Ansätze/Methoden direkt anhand einer klar definierten und quantitativ evaluierten Aufgabe verglichen wird, sind unshared tasks offen für verschiedenartige Beiträge, die auf einer gemeinsamen Datensammlung basieren. Shared und Unshared Tasks in den Digital Humanities sind ein vielversprechender Weg, Kollaboration und Interaktion zwischen Geistes-, Sozial- und ComputerwissenschaftlerInnen zu fördern und zu pflegen. Konkret riefen wir dazu auf, gemeinsam an einem heterogenen Korpus zu arbeiten, in dem Entitätenreferenzen annotiert wurden. Das Korpus besteht aus Parlamentsdebatten des Deutschen Bundestags, Briefen aus Goethes Die Leiden des jungen Werther, einem Abschnitt aus Adornos Ästhetischer Theorie und den Büchern von Wolframs von Eschenbach Parzival (mittelhochdeutsch). Auch wenn jede Textsorte ihre eigenen Besonderheiten hat, wurden alle nach einheitlichen Annotationsrichtlinien annotiert, die wir auch zur Diskussion stellten. Wir veröffentlichen hier den Aufruf zu Workshop-Beiträgen, die Annotationsrichtlinien, die Korpusdaten samt Beschreibung und die einführenden Vortragsfolien des Workshops.
  • Thumbnail Image
    ItemOpen Access
    Detecting protagonists in German plays around 1800 as a classification task
    (2018) Reiter, Nils; Krautter, Benjamin; Pagel, Janis; Willand, Marcus
    In this paper, we aim at identifying protagonists in plays automatically. To this end, we train a classifier using various features and investigate the importance of each feature. A challenging aspect here is that the number of spoken words for a character is a very strong baseline. We can show, however, that a) the stage presence of characters and b) topics used in their speech can help to detect protagonists even above the baseline.
  • Thumbnail Image
    ItemOpen Access
    A shared task for the digital humanities: annotating narrative levels
    (2019) Gius, Evelyn; Reiter, Nils; Willand, Marcus
    This volume is the first of two, and it documents activities that we have been conducting in the past years. They are best described as “organizing shared tasks with/in/for the digital humanities” and have evolved significantly since we started. Research in digital humanities entails a number of unique challenges, some of which are caused by the collaboration model that digital humanities projects often work in. This collaboration creates a lot of friction, but comes with huge potential: Different collaboration partners can specialize in different aspects of the shared goal. While a shared goal and a common language is still needed, each party does not have to know everything about the other party’s focus area. A (proto-)typical division of labor lends itself to the digital humanities: Computer scientists work on the technical aspects, while humanities scholars focus on the content side. Still, it turns out that knowing what each party is working on does not suffice, because content and technology need to be re-integrated at some point. How exactly this integration takes place depends on the specific project design and is a matter of interface.
  • Thumbnail Image
    ItemOpen Access
    Reading data: on digital reception studies
    (2018) Willand, Marcus; Beck, Jens; Reiter, Nils
    In the paper we present a method for the analysis of entity associations that real readers make in their reviews on goodreads.com, a social reading platform - and first results and insights of our analysis.
  • Thumbnail Image
    ItemOpen Access
    An interface to view collections of visual art
    (2023) Pflüger, Hermann; Viehhauser, Gabriel; Tumanov, Rostislav
    Art experts prefer being able to look at the individual images they are working on in the course of their research. However, if one were to look at digitally accessible images in the field of visual art, one would be dealing with billions of images; no one can handle visually examining such huge numbers of images one at a time. Therefore, art experts need special tools to examine and describe artworks in the context of other artworks. We used our experience from previous projects and interviews with members of the target group (art historians, curators, art dealers, and artists) to identify the central issues these experts encounter when working with large image collections and to determine the functionality and properties a system must offer to support their work. The results led to the customized interface LadeCA.View, which is now used in several projects. LadeCA.View enables experts to describe an exhibition or a collection of visual art in such a way that a user can obtain an overview of the intention, content, and structures of the exhibition or collection within a short period of time without looking at each image individually. LadeCA.View can also be used as an interface to probe more deeply into a collection or exhibition. In this paper we show the functions and visualizations of the interface and explain the design decisions. Furthermore, we outline LadeCA.View’s scope of applicability using three case studies