Options
A Language Model Sensitive to Discourse Context
Abstract
The paper proposes a meta language model that can dynamically incorporate the influence of wider discourse context. The model provides a conditional probability in forms of P (text|context), where the context can be arbitrary length of text, and is used to influence the probability distribution over documents. A preliminary evaluation using a 3-gram model as the base language model shows significant reductions in perplexity by incorporating discourse context.
Publikationstyp
ConferencePaper
Autor*in •
Noh, Tae-Gil
Padó, Sebastian
Erscheinungsdatum
2014
Fachbereich
Institut / Einrichtung
Erschienen in
Proceedings of the 12th edition of the KONVENS conference
Erste Seite
201
Letzte Seite
206
URN
urn:nbn:de:gbv:hil2-opus-2822
HilPub Permalink
Dateien p009.pdf (244.48 KB)
Main Conference Proceedings of the 12th Konvens 2014