On Mono- and Multilingual Word embeddings for NLP
Aus International Center for Computational Logic
On Mono- and Multilingual Word embeddings for NLP
Verteidigung Studienarbeit von Anatoly Zelenin
- Veranstaltungsort: APB 3027
- Beginn: 25. Februar 2016 um 13:30
- Ende: 25. Februar 2016 um 14:30
- Forschungsgruppe: Computational Logic
- Event series: KBS Seminar
- iCal
A fundamental problem of Natural Language Processing is to find a good numerical representation of words to be used in further processing steps. We present different recent mono- and bilingual approaches to map words into vector spaces. Knowing them we focus on issues of existing Multilingual approaches and discuss how to improve them further to build the foundation for systems analyzing Social Media posts from all over the world.
