BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Frontiers in Named Entity Recognition and Linking  -  Leon Derczyn
 ski and Kalina Bontcheva\, University of Sheffield
DTSTART:20150123T120000Z
DTEND:20150123T130000Z
UID:TALK53855@talks.cam.ac.uk
CONTACT:Tamara Polajnar
DESCRIPTION:Recognising mentions of entities in texts\, and grounding thes
 e into entries in external knowledge bases (e.g. DBpedia)\, provides a mea
 ns of describing document semantics and enabling search and indexing for s
 pecific concepts and ideas. These tasks\, referred to as Named Entity Reco
 gnition and Named Entity Linking\, have been well-studied in many text dom
 ains and genres. In the modern text processing environment\, in contrast t
 o dealing with named entities in formal or heavily structured documents su
 ch as newswire and official reports\, we now deal with very large volumes 
 of uncurated text from sources such as social media\, which poses broad an
 d challenging problems in NLP research.\n\nIn this talk\, we will discuss 
 current issues in social media NLP\, and report on new research in both na
 med entity recognition and linking in this genre\, detailing specific prob
 lem areas and demonstrating new solutions.\n
LOCATION:FW26\, Computer Laboratory
END:VEVENT
END:VCALENDAR