NLP Zurich:

Meetup.com: [ Ссылка ]
Linkedin: [ Ссылка ]
Twitter: [ Ссылка ]



Talk summary:
Taxonomies - classifications of knowledge - are a basic building block of our mental world model. As such, they have also been widely used in the NLP domain. However, the classical process of taxonomy construction is manual, cumbersome and doesn’t scale well in our world which is marked by ever-increasing change. The need for automated taxonomy construction has already been recognised for a while, but so far, related research is still experimental and fragmented.

In this talk, I will first explain the notion of a taxonomy from the cognitive and the linguistic point of view. Then, I provide an overview over existing methods of automatic taxonomy construction, discussing the pros and cons of these methods in the context of specific use cases. Finally, I present our own, hybrid approach that combines linguistic knowledge with embedding methods and demonstrate its application at the example of technological domains with a fast pace of change, such as AI and cybersecurity.

About the speaker:
Dr. Janna Lipenkova (www.jannalipenkova.com) holds a PhD in Computational Linguistics from the Free University in Berlin. Currently, she runs two analytics companies - Anacode and Equintel - and focusses on distilling hidden knowledge from unstructured data, thus helping businesses to generate a unique information advantage.

свернуть