Natural Language Processing

Master of Arts

MediaWiki based dictionary for minority languages

Nowadays, I am developing new methods for automatically enriching dictionaries for small Uralic languages. This is done in collaboration with the Language Bank of Finland. Stay tuned for more information in the future! Explore the dictionary.

Computational Creativity and NLG

I have worked for the research group of computational creativity at the University of Helsinki. There, I created a tool called Poem Machine. It boasts with an AI capable of creativity in the form of poetry. It makes it possible for humans to create poems together with a machine. Poem Machine is also featured on the Finnish news.

An important part for generating poems automatically, in my opinion, is to have a linguistically solid system. I.e. a tool that can create syntactically correct sentences from given words. That’s why I started to develop my very own Python library, syntaxmaker. It’s an NLG (Natural Language Generation) tool for Finnish. A quick start guide can be found at my personal GitHub.

Poem Machine also knows about the semantics of Finnish. Its data can partially be browsed through SemFi (The Finnish Semantic Data Explorer). And the full database can be downloaded by following the instructions on SemFi GitHub.

Automatic sarcasm detection

My master’s thesis for Spanish philology was about automatic sarcasm recognition in transcriptions of some South Park and Archer episodes.

The whole MA thesis is available in Spanish at the eThesis system of the University of Helsinki.