Finding expertise using online community dialogue and the Duality of Expertise

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

Abstract

The Duality of Expertise considers the “Expert” to be a social role dependent on an individual’s expertise claims and the opinion of their community towards those claims. These are the internal and external aspects of a person’s expertise. My Expertise Model incorporates this duality in a process designed for expertise finding software in an online community forum. In this model, a posting’s term usage is evidence of expertise claims. The dialogue acts in replies to those postings are evidence of the community’s opinion. The model’s preprocessing element uses a bricolage of linguistic and IR tools and methods in a novel way to construct each author’s expertise profile. For any topic query, the profiles are ranked to determine the Community Topic Expert. A series of experiments demonstrate the advantage of utilising the Duality of Expertise when ranking experts rather than just the internal or external aspects of expertise.
Original languageEnglish
Title of host publicationAustralasian Language Technology Association Workshop 2014 - Proceedings of the Workshop
Subtitle of host publication26 - 28th of November, 2014 RMIT Melbourne, Australia
EditorsGabriela Ferraro, Stephen Wan
Place of PublicationAustralia
PublisherAustralian Language Technology Association (ALTA)
Pages69-78
Number of pages10
Volume12
Publication statusPublished - 2014
EventAustralasian Language Technology Association Workshop 2014 - RMIT, Melbourne, Australia
Duration: 26 Nov 201428 Nov 2014
Conference number: 12th
https://www.aclweb.org/anthology/events/alta-2014/ (Proceedings)

Conference

ConferenceAustralasian Language Technology Association Workshop 2014
Abbreviated titleALTAW 2014
Country/TerritoryAustralia
CityMelbourne
Period26/11/1428/11/14
OtherALTA 2014 will be held in conjuction with the 19th Australasian Document Computing Symposium 2014 (ADCS 2014).
Internet address

Keywords

  • expertise
  • NLP

Cite this