BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Constructing topical hierarchies for Expertise Mining - Georgeta B
 ordea\, DERI\, NUI Galway
DTSTART:20140228T120000Z
DTEND:20140228T130000Z
UID:TALK50494@talks.cam.ac.uk
CONTACT:Tamara Polajnar
DESCRIPTION:In this age of pervasive internet access we have become accust
 omed to rely on web search for our most basic information needs. But compl
 ex queries in knowledge-intensive organisations\, as well as in the academ
 ic environment\, are best answered by direct interaction with domain exper
 ts. Current approaches for expert finding are based on keyphrase search\, 
 relying on exact string matches to identify experts. What is needed is sup
 port for exploratory search and discovery of  expertise topics and experts
 \, as well as content-based measures of expertise. Hierarchical structures
  have long been considered for content organisation and navigational suppo
 rt\, but in this talk we hypothesize that they also provide useful informa
 tion for measuring expertise.\n\nBuilding upon research for term extractio
 n and taxonomy construction\, we propose an Expertise Mining approach to e
 xpertise topic extraction\, expert finding and expert profiling. In partic
 ular\, we discuss a term extraction approach that considers the level of s
 pecificity of a term within a domain\, a graph-based algorithm for topical
  hierarchy construction using a global generality measure\, and a novel me
 asure of expertise based on topical hierarchies.\n\nThis work has been par
 t of the Saffron project [1]\, a system that provides insight into several
  Computer Science domains and which was deployed at different conferences 
 as a tool for finding collaborators.\n\n[1] http://saffron.deri.ie/
LOCATION:FW26\, Computer Laboratory
END:VEVENT
END:VCALENDAR