Members feature: Viktor Martinović

We are pleased to officially welcome Viktor Martinović from the Faculty of Philological and Cultural Studies to the research network Data Science @ Uni Vienna!

Viktor started his PhD in 2018 at the Finno-Ugric department at the University of Vienna. His research focuses on methodologies for the detection of ancient loanwords. As a researcher at the Department of Linguistic and Cultural Evolution at the Max Planck Institute for Evolutionary Anthropology in Leipzig, he worked on improvements in data standardisation and quantitative approaches in historical linguistics. He joined the Natural Language Processing (NLP) working group at the University of Vienna in 2024.


Why did you join the research network?

Throughout my research journey, I have met and worked within very diverse frameworks. The Finno-Ugric department, where I started my PhD, is a traditional humanities department, with an emphasis on qualitative approaches. The MPI-EVA in Leipzig is the polar opposite to that paradigm, where Bayesian statistics and algorithmic approaches dominate. Then there is computational linguistics, a classical engineering discipline, where benchmarks and evaluation metrics matter the most. These research communities are driven by different questions, ambitions, fears, and hopes. Over the years, I had to learn to navigate each environment. By joining this research network, I hope to learn more about its ways of working and would like to share from the experiences I have made on my journey so far.

How does your research or work relate to Data Science?

Initially, I wanted to check if there could be any ancient Gothic loanwords in Hungarian that were simply overlooked by researchers, similar to the Low German loanwords in Finnish that were overlooked until Mikko Bentlin’s PhD-thesis in 2007. However, I soon came to realize that despite ancient loanwords being quite the hot topic in Finno-Ugric studies, there is in fact no methodology for their detection. So, I started tinkering around with Python scripts to reverse-engineer a detection-pipeline based on results in the existing literature. And that is how I ended up in data science.

How would you like to contribute to the community?

I would say that my main quality is communication and connecting people. During my one-year research stays in Paris, Helsinki, Reykjavík and Leipzig, I had the chance to collaborate and make friends with people from various fields. Thus, I hope to contribute new connections and flows of ideas to the community.


Personal website: https://www.linkedin.com/in/martino-vic