Mark Granroth-Wilding

Current Research

My research interests lie in natural language processing (NLP), computational creativity and music processing/cognition.

Particular areas of research that I am currently focussing on are:

  • multilingual NLP and language transfer for low-resource languages;
  • application of NLP to noisy, historical data;
  • cross-lingual representation learning;
  • combination of Bayesian modelling and neural representation learning.
More

I am a postdoctoral research associate in the Discovery Group at the University of Helsinki, working with Hannu Toivonen. I work on the following projects:

  • NewsEye
    Building a digital investigator for historical newspapers. Topic modelling for analysis of multilingual, noisy output from automated text recognition.
  • Embeddia
    Addressing the challenges of multilinguality in the EU using cross-lingual embeddings to allow existing monolingual resources to be used across languages. Multilingual techniques for news analysis and media production.

Teaching

Upcoming! Mar–Apr 2021: Masters-level lecture course, Natural Language Processing.

Jan–Feb 2020: Masters-level lecture course, Natural Language Processing.

Nov–Dec 2019: One-off seminar course, Natural Language Processing and News.

May 2019: New masters-level intensive course, Introduction to Natural Language Processing.

Jan–Feb 2017: One-off seminar course (study group), Natural Language Generation for News Automation.

History

Earlier in my time as a postdoc in the Discovery Group, I worked on the project Digital Language Typology, discovering family relationships between languages with digital textual and speech material, focussing on low-resourced Uralic languages, funded by the Academy of Finland. I also worked on Immersive Automation, investigating and building tools for future news technology, including Natural Language Generation.

From 2013 to 2016, at the University of Cambridge, I worked on the What-If Machine (WHIM) EU FP7 project with Stephen Clark. It concerned building a computational model of creativity, constructing a system that automatically produces ideas that may serve as the basis for cultural artefacts. More...

From 2009 to 2013, I completed my PhD at the University of Edinburgh with Mark Steedman in 2013. The project concerned the application of NLP techniques to music processing. We developed a grammar to model the structure of tonal harmony and used statistical parsing for automatic harmonic analysis. More...

In 2008-9, I worked for Mercurytide.

Earlier, I did a masters in Informatics at the University of Edinburgh and an undergraduate degree in Computer Science at the University of Cambridge.