Bibliographie
Call for sentiment analysis systems to participate in WASSA
HACKATHON
BACKGROUND
Research in automatic Subjectivity and Sentiment Analysis (SSA), as
subtasks of Affective Computing and Natural Language Processing (NLP),
has flourished in the past years. The growth in interest in these tasks
was motivated by the birth and rapid expansion of the Social Web that
made it possible for people all over the world to share, comment or
consult content on any given topic. In this context, opinions,
sentiments and emotions expressed in Social Media texts have been shown
to have a high influence on the social and economic behaviour worldwide.
SSA systems are highly relevant to many real-world applications (e.g.
marketing, eGovernance, business intelligent, social analysis) and also
many tasks in NLP – information extraction, question answering, textual
entailment, to name just a few.
The importance of this field has been proven by the high number of
approaches proposed in research in the past decade, as well as by the
interest that it raised from other disciplines (Economics, Sociology,
Psychology, Marketing, Crisis Management, Behavioral Studies) and the
applications that were created using its technology.
In spite of the growing body of research in the area in the past years,
dealing with affective phenomena in text has proven to be a complex,
interdisciplinary problem that remains far from being solved. Its
challenges include the need to address the issue from different
perspectives and at different levels, depending on the characteristics
of the textual genre, the language(s) treated and the final application
for which the analysis is done. Additionally, SSA from Social Media
texts has opened the way to many other types of analyses, linking
textual data with images, social network metadata and the specific text
markings (e.g. Twitter hashtags) and giving the unprecedented
possibility to follow events unfolding through Twitter messages or
Facebook posts.
Finally, the possibility to follow trends on opinions, while comparing
and contrasting different sources of information (e.g. mainstream media
vs. social media) allows for a balanced, unbiased,
Sentiment analysis systems hackathon
In the light of the fact that different sentiment analysis systems have
been proposed and showcased in the past years, we feel there is a
growing need to make other researchers and users familiar with these
systems and have them employ them for building an end application.
The Hackathon word stands for “Hacking Marathon”, and its purpose is to
introduce some technology or software toolkit to the attendees, and let
then “play” and develop ideas around it.
We invite submissions for descriptions of sentiment analysis systems
that can used during the hackathon.
We envisage providing the systems with different types of data, in
different languages and from different domains. Subsequently, the
sentiment analysis systems participating in the hackathon will process
these data for the different applications envisaged by the users.
The following data will be provided for training and “playing
around”/evaluation :
- Eupean Media Monitor (http://emm.newsbrief.eu/overview.html) data with
mentions of entities in news (reported speech, quotations) -
multilingual
- annotated sentiment data from Twitter (English, German, ish,
Italian, French)
- annotated sentiment data from Facebook (English, German, ish,
Italian, French)
- OpeNER news : sentences from news articles annotated with opinion
holders, expressions and targets (Dutch, German, English, French)
- OpeNER hotel reviews annotated with opinion holders, expressions and
targets (Dutch, English, French, ish, Italian, German)
Organizers
*_Alexandra Balahur_**//*
*/European Commission Joint Research Centre/*
*/Institute for the Protection and Security of the Citizen (IPSC)/*
*/Via E. Fermi 2749, T.P. 267, 21027 Ispra (VA), Italy/*
*/alexandra.balahur@jrc.ec.europa.eu/*
*_Ruben Izquierdo_**//*
*/Vrije Universiteit (VU) Amsterdam/*
*/De Boelelaan 1105, level 13 /*
*/1081 HV Amsterdam, The Netherlands/*
*/ruben.izquierdobevia@vu.nl/*
*_Andrés Montoyo_*
*/University of Alicante, /*
*/Departamento de Lenguajes y Sistemas Informaticos,/*
*/Ap. De Correos 99, 03080 Alicante, Spain/*
*/montoyo@dlsi.ua.es/*
*_Isa Maks_*
*/Vrije Universiteit (VU) Amsterdam/*
*/De Boelelaan 1105, level 13 /*
*/1081 HV Amsterdam, The Netherlands/*
*/e.maks@let.vu.nl/*
*_Erik van der Goot_*
*/European Commission Joint Research Centre/*
*/Institute for the Protection and Security of the Citizen (IPSC)/*
*/Via E. Fermi 2749, T.P. 267, 21027 Ispra (VA), Italy/*
*/erik.van-der-goot@jrc.ec.europa.eu/*
IMPORTANT DATES
- System description submission/demo paper : March 25, 2016
- Training data available : 31 March 2016
- Hackathon to at NAACL-HLT 2016 : 16 June 2016
SUBMISSIONS
The system descriptions can also be submitted as demo papers.
They must not exceed five (5) pages without references.
The paper should make clear what domain, language and genre the system
is best for
Papers for WASSA should be submitted using the NAACL-HLT 2016 Style
Files, available at : http://naacl.org/naacl-hlt-2016/cfp.html
Demo papers will be published in the NAACL HLT WASSA proceedings,
included in the ACL Anthology. The best papers will be chosen for a
special issue of an ISI- indexed journal. Previous special issues of
WASSA were/are in the process of being published in the Decision Support
Systems, Computer Speech and Language and Information Processing and
Management journals (Elsevier).
To submit a paper, please access :
https://www.softconf.com/naacl2016/WASSA2016/
Looking forward to your contributions !
All the best,
The WASSA 2016 Organizers