University of Cambridge > Talks.cam > NLIP Seminar Series > Generation of Referring Expressions: Evaluating some standard algorithms

Generation of Referring Expressions: Evaluating some standard algorithms

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact NLIP Seminars.

In this seminar some ongoing work will be presented that is carried out in the EPSRC -funded TUNA project (Towards a UNified Algorithm for Generating Referring Expressions). The aims of this work are to build a semantically transparent corpus of referring expressions and to use this corpus to evaluate some of the best-known algorithms for the generation of referring expressions. The seminar, which reports on joint work with Kees van Deemter and Albert Gatt, will focus on three issues: (1) the production experiment that we used to gather the data for the corpus, (2) the annotation of the corpus data and (3) the evaluation of the generation algorithms, which is done by comparing the algorithms’ output with the referring expressions in the corpus.

Background reading: *http://www.csd.abdn.ac.uk/~ivdsluis/publications/inlg2006_deemter_sluis_gatt.pdf

This talk is part of the NLIP Seminar Series series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2019 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity