I stumbled upon the field of Natural Language Processing while studying Computer Science. Almost immediately, I felt at home and decided to shift my focus. What followed was a Joint Bachelor in Computer Sciences and Digital Philology, as well as a Masters in Linguistic and Literary Computing, for which I am currently completing my thesis, both at TU Darmstadt, in Germany. A particular penchant of mine is to dive deep into subjects that are the foundation of the field, such as tokenisation and other preprocessing steps. After all, every machine learning model can only be as good as its underlying data allows it to be.
In my free time, I like to read an unhealthy number of books, while consuming a similar amount of tea. While I enjoy cooking, what really interests me is poring over old recipe books, most of which desperately want to be rescued from obscurity by applying Optical Character Recognition (OCR) to scans readily available online. This is simple enough for the English language, though far from easy, but considerably harder for German. If you share similar interests, do not hesitate to get in touch with suggestions, reading or otherwise.
blog@unnlp.com
This blog does not track you in any way, which explains the blissful absence of any cookie banners.
This blog is based on the Hyde theme (v2.1.0) for Poole, which is sadly unsupported at the time of writing (2021), running on Github Pages but with some modifications, mostly by frederikaverpil. What follows is a non-exhaustive list of personal alterations to the design and site:
The source code for this blog is available here.