Adaptive Language Technologies : Informatisches Kolloquium

Link:
  • https://lecture2go.uni-hamburg.de/l2go/-/get/v/21677
Autor/in:
Beteiligte Personen:
  • Regionales Rechenzentrum der Universität Hamburg/ MCC/ Lecture2Go
  • eLB-MIN
Verlag/Körperschaft:
Universität Hamburg
Erscheinungsjahr:
2017
Medientyp:
Audiovisuell
Schlagwort:
  • Informatik
Beschreibung:
  • Automatic natural language understanding enables natural communication with computers and computer-assisted access to the content of large document collections. While classical approaches to artificial intelligence anticipate all possible situations and interactions in form of a fully specified dialogue model or ontology, they are hard to adapt to new domains and do not cope well with language change. In this talk, I will motivate an adaptive, purely data-driven approach to natural language processing. Illustrated by recent research prototypes, three stages of data-driven adaptation will be illustrated: feature/resource induction, induction of processing components and continuous data-driven learning. Finally, I will discuss current research and future directions regarding the integration of symbolic and statistical knowledge, interpretability of language processing components as well as advanced forms of information access.
Beziehungen:
URL https://lecture2go.uni-hamburg.de/l2go/-/get/l/5013
Lizenz:
  • UHH-L2G
Quellsystem:
Lecture2Go UHH

Interne Metadaten
Quelldatensatz
oai:lecture2go.uni-hamburg.de:21677