Emotions are an important part of our everyday lives. However, until quite recently, human
affect was regarded in a dual manner - positively, for its regulatory power and negatively, as a
sort of a "weakness" of the human spirit, that should ideally be rational, logical, *thinking* in a
very matter of fact and consequence-based fashion.
Recent discoveries in Neuropsychology and the possibilities opened by the functional Magnetic
Resonance Imaging have made it clear that emotions play a very important role for the
well-functioning of the human body, both psychologically, as well as physically.
Apart from the importance emotions have for each human being individually, research in
Social Psychology and disciplines such as Marketing, Mass-media Communication or Political
Science, has shown time and time again that the emotional discourse, its content - in words
with affective connotation and the combination thereof - is of paramount difference between the
success and failure of social actions, consumer products or political candidates.
Given that nowadays messages with (sometimes) high emotional connotations are so easily
shared using Social Media platforms and that their high volume makes manual sifting mostly impossible,
the automatic processing of Subjectivity, Sentiment and Emotions in texts, especially
in Social Media contexts is highly relevant.
Bearing these observations in mind, the aim of the 6th Workshop on Computational Approaches
to Subjectivity, Sentiment and Social Media Analysis (WASSA 2015) was to continue
the line of the previous editions, bringing together researchers in Computational Linguistics
working on Subjectivity and Sentiment Analysis and researchers working on interdisciplinary
aspects of affect computation from text. Starting with 2013, WASSA has extended its scope and
focus to Social Media phenomena and the impact of affect-related phenomena in this context.
The past two editions have shown important breakthroughs in dealing with the challenges of
these types of texts, in monolingual, multilingual and cross-domain contexts.
WASSA 2015 was organized in conjunction to EMNLP 2015: the Conference on Empirical
Methods in Natural Language Processing, on September 17, 2015, in Lisboa, Portugal.
For this year’s edition of WASSA, we received a total of 48 submissions, from universities
and research centers all over the world, out of which 8 were accepted as long and another 15 as
short papers. Each paper has been thoroughly reviewed by at least 2 members of the Program
Committee. The accepted papers were all highly assessed by the reviewers, the best paper receiving
an average punctuation (computed as an average of all criteria used to assess the papers)
of 4.8 out of 5.
BALAHUR-DOBRESCU Alexandra;
VAN DER GOOT Erik;
VOSSEN Piek;
MONTOYO Andres;
2017-01-16
The Association for Computational Linguistics
JRC97442
978-1-941643-32-7,
http://www.emnlp2015.org/proceedings/WASSA/WASSA-2015.pdf,
https://publications.jrc.ec.europa.eu/repository/handle/JRC97442,
