Summer School “Computational Language Technologies for Medievalists”

Graz, 8-12 July 2024

Embark on a five-day journey hosted by the University of Graz:


Natural Language Processing (NLP) has emerged as a crucial skill in many Digital Humanities scenarios seeking to unlock the potential of language-based technologies. Due to the particularities of medieval languages (e.g. special characters, historical language levels, scarce training data) medieval studies put particular challenges to the standard NLP methods. In our Summer School, we aim to guide postgraduate and PhD level participants through the conceptual foundations of NLP, offering hands-on exercises tailored to empower medievalists in curating content and managing metadata. This will include training in topic modeling, text re-use, authorship attribution, stylometry and basic application of large language models. Discover how to optimize your time and efforts effectively while integrating new tools into your scholarly pursuits.

In preparation for the main curriculum, we organize a Python crash course specifically designed to support novice applicants. These materials will provide foundational knowledge and essential skills, ensuring that participants, regardless of their level of expertise, can fully engage with and benefit from the Summer School experience.

No prior advanced mathematics or computer science knowledge is mandatory. Possessing basic Python syntax knowledge and command line-based or interactive computing environment-based (Jupyter Notebook, Google Colab) script execution is, however, advantageous. We expect basic computer literacy from all participants. You can check this via here!


How to Apply

To be considered for participation in this Summer School, please submit a one-page CV and a concise letter of interest (maximum one page) addressing the following:

  •    Why do you wish to attend this Summer School? Submissions describing a concrete use case you consider (data, research questions) will have preference. You will get the chance to present your research in a poster session.
  •    Share your academic background and research specialization.
  •    Sketch your technology skills in the above mentioned fields


This initiative is generously funded by the University of Graz, ZIM-ACDH and the ERC DiDip Project.

Kindly email your application material to didip[at]uni-graz.at with the subject “Application for NLP Summer School“. The deadline for applications is 15 March 2024. Applicants will be notified of acceptance by 15 April 2024.

Notice: Application is closed!

Program

Please note that the program may be under further consideration!

Objectives: The Summer School on Natural Language Processing for Medieval Texts is designed to equip students and scholars at graduate, doctoral, and postdoctoral levels with essential skills in Natural Language Processing (NLP) and its applications in medieval studies. The program offers a hands-on approach, providing practical tools and resources to enhance research in this specialized area. To foster a collaborative and enriching learning environment, the summer school will select 20+2 participants who will be divided into four groups, mixing individuals with varying levels of Python programming knowledge. This team-based approach is aimed at optimizing the learning curve by encouraging peer learning and leveraging diverse skill sets within each group, thus enhancing the overall educational experience for all participants.

Daily Schedule

  • 09:00-10:30 — Morning Session
  • 10:30-11:00 — Coffee Break
  • 11:00-12:30 — Mid-Morning Session
  • 12:30-13:30 — Lunch Break
  • 13:30-15:00 — Afternoon Session
  • 15:00-15:30 — Coffee Break
  • 15:30-17:00 — Late Afternoon Session
  • 17:00-17:30 — Coffee Break
  • 17:30-19:00 — Special Event or Keynote
  • Evening — Reception or Social Activity

Day 1: Introduction to NLP and Text Analysis

09:00-10:30 Lucija Krusic (Univ. Graz): Python programming course for novice participant (technical)

Obligatory for Novice participants

This course introduces novice participants to the fundamentals of Python programming, covering topics such as data types, control structures, functions, and basic libraries. Participants will learn to write simple Python scripts and programs, enabling them to solve NLP-related problems through hands-on projects and exercises.

10:30-11:00 Coffee Break

11:00-12:30 Klara Venglarova (Univ. Graz): Introduction to NLP Methods

Obligatory for Novice participants

What are types and tokens? What role does a lemmatizer play? How can computers recognize names in text? This introductory course on Natural Language Processing (NLP) covers fundamental concepts and includes practical exercises to deepen your understanding.

12:30-13:30 Lunch Break

13:30-15:00 Stefanie Dipper (Ruhr-Universität Bochum): Corpus Building

This session makes introductions to corpora and annotations, including the importance of various annotation types and methodologies for enhancing linguistic research through automated tools, with a focus on enhancing the usability and accuracy of linguistic data.

15:00-15:30 Coffee Break

15:30-17:00 Tamás Kovács (Univ. Graz): Introduction into our Working Environment (Colab, Data sets)

This introductory session will familiarize participants with our primary working environment, Google Colab, including how to access and manipulate datasets within this cloud-based platform. We will explore basic features of Colab notebooks, discuss data importation techniques, and demonstrate effective ways to interact with datasets to kickstart data-driven projects.

17:00-17:30 Coffee Break

17:30 – Jean-Baptiste Camps (École Nationale des Chartes): Computational Medieval Philology(Keynote)

This keynote will delve into the intersection of computational methods and medieval philology, demonstrating how digital tools and algorithms can enhance the analysis and understanding of ancient texts.

Reception

Day 2: Topic Modeling

09:00-10:30 Gabriel Viehhauser (Univ. Vienna): Topic Modeling

with Martina Scholger,Teaching Assistants: Roman Bleier, Niklas Tscherne, Nicolas Renet, Florian Atzenhofer-Baumgartner

This segment of the course will delve into topic modeling, a method used to uncover hidden thematic structures in large text corpora. Participants will learn about different algorithms such as Latent Dirichlet Allocation (LDA) and Non-negative Matrix Factorization (NMF), and apply these techniques to real datasets to identify and interpret significant topics.

10:30-11:00 Coffee Break

11:00-12:30 Mid-Morning Session

12:30-13:30 Lunch Break

13:30-15:00 Hands on with prepared task

15:00-15:30 Coffee Break

15:30-17:00 Hands on with own material

17:00-17:30 Coffee Break

17:30 – Poster Session

Day 3: Named Entity Recognition

09:00-10:30 Ismail Prada (Univ. Bern): Named Entity Recognition

with Sandy Aoun and Selina Galka, Teaching Assistants: Roman Bleier, Florian Atzenhofer-Baumgartner, Niklas Tscherne, Nicolas Renet, Klara Venglarova

This session will explore Named Entity Recognition (NER), a crucial technique in natural language processing that identifies and classifies named entities in text into predefined categories such as persons, organizations, locations, and others. Participants will learn to implement and train NER models using popular libraries like spaCy and NLTK, and apply these models to extract structured information from unstructured text data.

10:30-11:00 Coffee Break

11:00-12:30 Mid-Morning Session

12:30-13:30 Lunch Break

13:30-15:00 Hands on with prepared task

15:00-15:30 Coffee Break

15:30-17:00 Hands on with own material

17:00-17:30 Coffee Break

17:30 – Poster Session

Social Dinner

Day 4: Text Reuse and Authorship Analysis

09:00-10:30 Jeroen de Gussem (Univ. Gent): Text Reuse / Stylometry

with Tamás Kovács, Teaching Assistants: Bernhard Bauer, Florian Atzenhofer-Baumgartner, Niklas Tscherne, Nicolas Renet, Klara Venglarova

This course module will cover text reuse and authorship analysis, focusing on techniques to detect instances of text borrowing and to determine authorship in various texts. Students will explore algorithms such as cosine similarity, Jaccard index, and stylometric analysis, applying these methods to literary works and academic papers to identify patterns of reuse and attribute texts to their likely authors.

10:30-11:00 Coffee Break

11:00-12:30 Mid-Morning Session

12:30-13:30 Lunch Break

13:30-15:00 Hands on with prepared task

15:00-15:30 Coffee Break

15:30-17:00 Hands on with own material

17:00-17:30 Coffee Break

17:30 – Poster Session

City Tour

Day 5: Future Directions (LLM)

09:00-10:30 Axel Pichler (Univ. Stuttgart): LLM

This concluding session will discuss the future directions of language models, particularly large language models (LLMs), emphasizing their evolving capabilities, ethical considerations, and potential impacts on various industries. Participants will explore cutting-edge developments, such as generative adversarial networks and reinforcement learning from human feedback, and discuss how these innovations could shape the next generation of AI applications.

10:30-11:00 Coffee Break

11:00-12:30 Mid-Morning Session

12:30-13:30 Lunch Break

Excursion Buschenschank: A Buschenschank is a type of seasonal wine tavern in Austria, usually run by winemakers. They are known for serving their own wine and simple, locally sourced food, often in a rustic setting.


Thanks to the support of the university network HFDT we can offer 10 selected applicants financial support for accommodation in Graz.

https://static.uni-graz.at/fileadmin/_processed_/4/3/csm_bf_hfdt_qr_2d27b09bab.png

OpenEdition suggests that you cite this post as follows:
ktamas (February 14, 2024). Summer School “Computational Language Technologies for Medievalists”. DiDip. Retrieved July 12, 2024 from https://didip.hypotheses.org/2480


You may also like...

Search OpenEdition Search

You will be redirected to OpenEdition Search