BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Natural Language Generation from Semantic Web Ontologies - Gerasim
 os Lampouras\, UCL
DTSTART:20150605T110000Z
DTEND:20150605T120000Z
UID:TALK59287@talks.cam.ac.uk
CONTACT:Tamara Polajnar
DESCRIPTION:The Semantic Web is an effort to establish standards and mecha
 nisms that allow computers to reason more easily about the semantics of We
 b resources (documents\, data etc.). Ontologies play a central role in thi
 s endeavour\; given an OWL ontology for a knowledge domain\, one can publi
 sh on the Web machine-readable data pertaining to that domain (e.g.\, cata
 logues of products\, their features etc.)\, with the data having formally 
 defined semantics based on the conceptualization of the ontology. Several 
 OWL syntaxes have been developed\, but people unfamiliar with formal knowl
 edge representations often have difficulties understanding them. \n\nThis 
 talk is organized in three parts\, with the first part discussing natural 
 language generation methods that allow end-users to view ontology-based kn
 owledge representations in the form of automatically generated texts in mu
 ltiple natural languages. Arguing against the pipeline architecture of mos
 t natural language generation systems\, where the stages are greedily cons
 idered one after the other\, the second part aims to further improve the q
 uality of generated texts by introducing an Integer Linear Programming mod
 el that jointly considers content selection\, lexicalization\, sentence ag
 gregation\, and a limited form of referring expression generation. Lastly\
 , in an attempt to reduce the tedious and time consuming manual authoring 
 of domain-dependent linguistic resources that NLG needs to produce high qu
 ality texts\, the last part examines text mining and machine learning meth
 ods to automatically or semi-automatically extract the most important of t
 he resources from the Web.
LOCATION:FW26\, Computer Laboratory
END:VEVENT
END:VCALENDAR