BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:NLIP Reading Group: Unsupervised Decomposition of a Document into 
 Authorial Components - Thomas Lippincott (University of Cambridge)
DTSTART:20111013T110000Z
DTEND:20111013T120000Z
UID:TALK33713@talks.cam.ac.uk
CONTACT:Jimme Jardine
DESCRIPTION:Tom will be presenting \n\n@inproceedings{koppel2011unsupervis
 ed\n\,	author	= {Koppel\, M. and Akiva\, N. and Dershowitz\, I. and Dersho
 witz\, N.}\n\,	title	= {Unsupervised decomposition of a document into auth
 orial components}\n\,	booktitle	= {Proceedings of the 49th Annual Meeting 
 of the Association for Computational Linguistics: Human Language Technolog
 ies-Volume 1}\n\,	year	= {2011}\n\,	pages	= {1356--1364}\n\,	organization	
 = {Association for Computational Linguistics}\n}\n\n*Abstract*\nWe propose
  a novel unsupervised method for separating out distinct authorial compone
 nts of a document. In particular\, we show that\, given a book artificiall
 y “munged” from two thematically similar biblical books\, we can separ
 ate out the two constituent books almost perfectly. This allows us to  aut
 omatically  recapitulate many conclusions reached by Bible scholars over c
 enturies of research. One of the key elements of our method is exploitatio
 n of differences in synonym choice by different authors.
LOCATION:GS15\, Computer Laboratory
END:VEVENT
END:VCALENDAR