Orient- und Islamwissenschaft

Dr. Maroussia Bednarkiewicz

Maroussia Bednarkiewicz is since September 2020 postdoctoral researcher at the Institute for Asian- and Oriental Studies of the University of Tübingen. She studied German, Russian and Translation at the University of Geneva, and Islamic Studies at the University of Oxford. In her doctoral dissertation she analysed the early history of the Islamic call to prayer within the Hadith literature. Her project in Tübingen in collaboration with the Cluster of Excellence for Machine Learning explores the development of new digital methods for the study of Hadith texts and Arabic literature in general.

 

Main project: Time, Space and Relations in Hadith Literature, a machine-learning approach

The aim of the present research project is to achieve a unique, broader perspective on the dynamics of narrative adaptation in ḥadīth literature by applying advanced data analysis. We have now digitized databases as well as sophisticated and efficient algorithms, which were the missing factors for large-scale studies of ḥadīth literature as a whole. Using comprehensive datasets and recent algorithmic advances in text processing and machine learning (ML) models, I want to map the use, reuse and adaptations of ḥadīth narratives to better apprehend regional and general patterns within the whole ḥadīth literature. These patterns will both illustrate the diversity of Muslim societies and the Islamic specificities that unite them.

A full description of the project is available in the link attached to the title of this blurb.

Isnalyser

Automated isnād tree visualisations

In collaboration with Stefan Wezel.

 

For scholars studying Ḥadīth texts, drawing an isnād tree with more than 40 transmitters is a tedious work and finding the right medium to desplay it fully can even prove to be impossible. The isnalyser is a simple program for the automation of isnād trees drawing and their customisable display in handy formats.

https://github.com/dhakarat/isnalyser

New Languages for NLP Princeton Fellowship

Building Linguistic Diversity in the Digital Humanities

I was recently awarded a fellowship with the Center for Digital Humanities at Princeton, in collaboration with Haverford College, the Library of Congress Labs, and DARIAH, the European Digital Research Infrastructure for the Arts and Humanities, to increase diversity in the field of Natural Language Processing (NLP). Guided by leaders in the fields of multilingual NLP and DH I will have the great priviledge of collaborating with Irene Kirchner and an inspiring cohort of scholars from around the world. In this intellectually stimulating environment, Irene and I will be enhancing the presence of classical Arabic on the NLP scene. For the annotation work we have been collaborating with Tulaib Zafir (AM candidate, Harvard University) and Nuh Elalaoui (MA candidate, Tübingen University).