Semi-supervised topic classification for low resource languages

In this paper, we present a novel methodology for rapidly developing a topic-based document classification system for a language that has limited resources. Our approach, a hybrid one, combines supervised and unsupervised topic classification techniques. Given that access to native speakers is fairl...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2008 IEEE International Conference on Acoustics, Speech and Signal Processing S. 5093 - 5096
Hauptverfasser: Daben Liu, McVeety, S., Prasad, R., Natarajan, P.
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.03.2008
Schlagworte:
Broadcasting
Hidden Markov Model
Hidden Markov models
Humans
Internet
Malay
Natural languages
off-topic rejection
Runtime
Search engines
Testing
topic clustering
Topology
unsupervised topic discovery
Web sites
ISBN:9781424414833, 1424414830
ISSN:1520-6149
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In this paper, we present a novel methodology for rapidly developing a topic-based document classification system for a language that has limited resources. Our approach, a hybrid one, combines supervised and unsupervised topic classification techniques. Given that access to native speakers is fairly limited for low resource languages, our approach requires annotating only a few broad "root" topics in the corpus. Next, unsupervised topic discovery (UTD) technique is used to automatically determine finer topics within the root topics. Lastly, we use the recently developed unsupervised topic clustering technique to organize the corpus into a hierarchical structure that enables browsing documents at multiple levels of granularity. Recognizing the need for reducing false alarms during runtime, we describe rejection techniques for discarding off-topic documents.
ISBN:9781424414833
1424414830
ISSN:1520-6149
DOI:10.1109/ICASSP.2008.4518804