[Corpora-List] Sentiment-annotated quotation corpus available for download

[Ralf Steinberger]

Readers of this list may be interested in knowing about a newly available
resource to work on sentiment analysis, or opinion mining. The resource can
be freely downloaded from 
 
     http://langtech.jrc.ec.europa.eu/JRC_Resources.html .
 
The Excel sheet contains a set of 1590 English language quotations (reported
speech) manually annotated by sets of two independent annotators for
sentiment (positive, negative, objective/neutral) expressed towards the
entities mentioned inside the quotation. The purpose of the annotation was
to produce a gold standard collection of news snippets. We believe that
opinion mining from the news is different compared to other text types such
as reviews.
 
Note that the sentiment value refers to the entity mentioned inside the
quotation, and not to the entire text of the quotation. Also, an attempt was
made to distinguish positive or negative sentiment from good or bad news
(see Balahur & Steinberger 2009 for details).
 
The quotations were identified using the algorithm and tool described in: 
 
Pouliquen Bruno, Ralf Steinberger & Clive Best (2007). Automatic Detection
of Quotations in Multilingual News. In: Proceedings of the International
Conference Recent Advances in Natural Language Processing (RANLP'2007), pp.
487-492. Borovets, Bulgaria, 27-29.09.2007.
 
For details on this set of quotations and for some experiments using the
test set, see the following publication. Please use this publication when
referring to the data set:
 
Balahur Alexandra, Ralf Steinberger, Mijail Kabadjov, Vanni Zavarella, Erik
van der Goot, Matina Halkia, Bruno Pouliquen & Jenya Belyaeva (2010).
Sentiment Analysis in the News. In: Proceedings of the 7th International
Conference on Language Resources and Evaluation (LREC'2010), pp. 2216-2220.
Valletta, Malta, 19-21 May 2010. Available from:
http://langtech.jrc.ec.europa.eu/Documents/2010_03_LREC_Sentiment-analysis.p
df 
 
USAGE CONDITIONS:
 
You can use and distribute this resource freely, but please make reference
to its authors (via the above-mentioned publication).
 
For the main motivation why we are aiming at separating positive and
negative sentiment from good or bad news, see the following paper:
 
Balahur-Dobrescu Alexandra & Ralf Steinberger (2009). Rethinking sentiment
analysis in the news: from theory to practice and back. 'Workshop on Opinion
Mining and Sentiment Analysis' (WOMSA), held at the 2009 CAEPIA-TTIA 13th
Conference of the Spanish Association for Artificial Intelligence, pp. 1-12.
Sevilla, Spain, 13.11.2009. Available from:
http://langtech.jrc.ec.europa.eu/Documents/09_WOMSA-WS-Sevilla_Sentiment-Def
_printed.pdf 
 
Alexandra Balahur, University of Alicante, Spain 
Ralf Steinberger, European Commission - Joint Research Centre (JRC), Italy 
 
 
Ralf Steinberger ( <http://langtech.jrc.ec.europa.eu/RS.html>
http://langtech.jrc.ec.europa.eu/RS.html) 
European Commission - Joint Research Centre (JRC)
URL - Applications:  <http://emm.jrc.it/overview.html>
http://emm.jrc.it/overview.html 
URL - The science behind them:  <http://langtech.jrc.ec.europa.eu>
http://langtech.jrc.ec.europa.eu 
21027 Ispra (VA), Italy


 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/corpora/attachments/20100609/26514cfc/attachment.htm>
-------------- next part --------------
_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora