UNIGE document Chapitre d'actes
previous document  unige:24683  next document
add to browser collection

Ontology-Based Information Extraction from Twitter

Published in Proceedings of the Workshop on Information Extraction and Entity Analytics on Social Media Data - COLING 2012. Mumbai (India) - December 2012 - The COLING 2012 Organizing Committee. 2012, p. 17-22
Abstract The popular microblogging service Twitter provides a vast amount of short messages that contains interesting information for Information Extraction tasks. This paper presents a rule-based system for the recognition and semantic disambiguation of named entities in tweets. As our experimental results shows, performance of this approach measured through BDM looks promising when using Linked Data as Freebase and syntactical context for disambiguation.
Keywords Ontology-based Information ExtractionDisambiguationTwitterLinked Data
Stable URL https://archive-ouverte.unige.ch/unige:24683
Full text
Research group Laboratoire d'Analyse et de Traitement du Langage (LATL)

816 hits



Deposited on : 2012-12-17

Export document
Format :
Citation style :