Automatic Summarization of Short Fiction
Anna Kazantseva (ankazant <at> site (point) uottawa (point) ca)
Wednesday 8 June 2011 at 11:30 AM
Salle 3195, Pavillon André-Aisenstadt
Most research on automatic text summarization revolves around genres with relatively well-defined structure: newswire, scientific papers, legal documents, etc. In this talk I will present an approach to summarizing short fiction – a relatively unexplored data type. The objective is a particular kind of summary – one that helps the reader decide whether she wants to read a story or not. A number of characteristics of literary texts make them a difficult target for summarization: the rate of lexical repetition is much lower, position of sentences in text is not indicative of how informative they are, the use of metaphor and figurative language is very wide-spread, to name just a few. Our approach relies on using two main types of information: the information about the main entities in stories and the information about the aspectual type of sentences. While the work is preliminary, the resulting summaries have been evaluated by 15 judges. The results suggest that the summaries are helpful for achieving the original objective.
To receive weekly talk announcements, please send an e-mail to email@example.com. Simply write a message containing the single line 'subscribe ralli' (without the quotes).
See all the weekly talks for the year:1991 1992 1993 1994 1995 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022