Knowledge base population for organization mentions in email

Ning Gao, Mark Dredze, Douglas W. Oard

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A prior study found that on average there are 6.3 named mentions of organizations found in email messages from the Enron collection, only about half of which could be linked to known entities inWikipedia (Gao et al., 2014). That suggests a need for collection-specific approaches to entity linking, similar to those have proven successful for person mentions. This paper describes a process for automatically constructing such a collection-specific knowledge base of organization entities for named mentions in Enron. A new public test collection for linking 130 mentions of organizations found in Enron email to either Wikipedia or to this new collection-specific knowledge base is also described. Together, Wikipedia entities plus the new collectionspecific knowledge base cover 83% of the 130 organization mentions, a 14% (absolute) improvement over the 69% that could be linked to Wikipedia alone.

Original languageEnglish (US)
Title of host publicationProceedings of the 5th Workshop on Automated Knowledge Base Construction, AKBC 2016 at the 2016 Conference of the North American Chapter of the Association for Computational Linguistics
Subtitle of host publicationHuman Language Technologies, NAACL-HLT 2016
EditorsJay Pujara, Tim Rocktaschel, Danqi Chen, Sameer Singh
PublisherAssociation for Computational Linguistics (ACL)
Pages24-28
Number of pages5
ISBN (Electronic)9781941643532
StatePublished - 2016
Event5th Workshop on Automated Knowledge Base Construction, AKBC 2016 at the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2016 - San Diego, United States
Duration: Jun 17 2016 → …

Publication series

NameProceedings of the 5th Workshop on Automated Knowledge Base Construction, AKBC 2016 at the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2016

Conference

Conference5th Workshop on Automated Knowledge Base Construction, AKBC 2016 at the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2016
Country/TerritoryUnited States
CitySan Diego
Period6/17/16 → …

ASJC Scopus subject areas

  • Computer Science Applications
  • Artificial Intelligence
  • Language and Linguistics
  • Linguistics and Language

Fingerprint

Dive into the research topics of 'Knowledge base population for organization mentions in email'. Together they form a unique fingerprint.

Cite this