Word embedding similarity is tricky
Even though word embeddings are widely used, we do not know a whole lot about them
May 6, 2024
I am a computational linguist interested in the mechanisms underpinning (human and machine) learning.
My tinkering is primarily done in 🐍, which I firmly believe to have one of the most productive standard libraries to date. Other than that most any problem deserves its own solution.
Presently I work for YUKKA Lab, a news analytics company, as the data engineering team lead of a fantastic group of people, where we are aiming to create a comprehensive knowledge graph of the world. Before I’ve worked with the great folks at yWorks and mecodia.
If you feel like my posts were useful to you, you can contact me under lbe at yukkalab dot com.