Veranstaltungsarchiv 2024

 

Dienstag, 12. März 2024
CLARIN Café: Computer-Assisted Pragmatic Annotation of Native and Learner Corpora

Uhrzeit: 14:00–16:00 Uhr

Ort: virtuell

Programm: https://www.clarin.eu/event/2024/clarin-cafe-computer-assisted-pragmatic-annotation-native-and-learner-corpora

 

Donnerstag, 21. März 2024
Mitgliederversammlung

Uhrzeit: 12:30–13:30 Uhr

Ort: Forschungsinstitut Brenner-Archiv

Programm:

  • Vortrag: Nicola Brocca/Joseph Wang: LadderWeb: An AI-based assistant for the pragmatic annotation of cancellations and requests
  • Vortrag: Tim Sieberichs: Neue Preismodelle von Transkribus
  • DH-Wahlpakete: Studienjahr 2024-25
  • Einrichtung eines Beirats FZDH
  • Stud. MA “Data Stewardship”
  • Allfälliges 

 

Donnerstag, 25. April 2024
Vorstellung von Omeka S & Projektpräsentationen

Uhrzeit: 12:30–13:30 Uhr

Ort: Agnes Heller Haus

Programm:

 

Donnerstag, 25. April 2024
DH-Zentrum Lunchtime Lecture & Online-Workshop

Datum & Uhrzeit: Donnerstag, 20. Juni 2024, 12:00–14:30 Uhr

Ort: Brenner-Archiv & Online

Programm:

  • Begrüßung
  • 12.00 Uhr: Kaspar Beelen: Source Criticism at Scale. Diagnosing Bias and Fairness in Digital Heritage

Abstract: Research in the Digital Humanities operates at scales often beyond the reach of qualitative methods. It involves vast amounts of data—billions of words or millions of images. But widening scope also presents novel challenges. Increasing scale distances scholars from their sources and raises issues of bias and fairness in their data and workflow. In this talk Kaspar Beelen will discuss methods for investigating hidden biases in large collections, focussing on newspaper collections but also discussing applications to digital heritage more broadly. He demonstrates how source criticism at scale allows scholars to explore questions of representativeness and bias in novel ways, analysing which voices are missing or under-represented in big heritage data.

  • Q & A, Diskussion
  • 13:15 Uhr: Hands-On Online-Workshop mit Kaspar Beelen: A gentle introduction to using open-source Large Language Models for humanities analysis

Abstract: In this interactive session, we start with a brief introduction to language models: what are these models, how are they created, and why are so many researchers currently excited (but also worried) about them? After explaining some basic concepts, we have a closer look at a few hands-on practical examples in which we use LLMs to interrogate historical data. In this session, you will learn the fundamental techniques to work with open-source LLMs such as Llama3 and experiment with different prompting strategies.

 

Freitag, 13. September 2024
DiSCourse - The Digital Science Seminar Series on: Information Extraction from Noisy Text Output

Datum & Uhrzeit: Freitag, 13. September 2024, 12:00 Uhr

Ort: Digital Science Center

weitere Informationen: https://www.uibk.ac.at/en/disc/events/discourse-doucet/


[zurück]

Nach oben scrollen