=Paper=
{{Paper
|id=Vol-1351/preface
|storemode=property
|title=None
|pdfUrl=https://ceur-ws.org/Vol-1351/preface.pdf
|volume=Vol-1351
}}
==None==
https://ceur-ws.org/Vol-1351/preface.pdf
ESSEM 2015: Emotion and Sentiment in Social
and Expressive Media
Cristina Bosco1 , Erik Cambria2 , Rossana Damiano1
Viviana Patti1 , Paolo Rosso3
1
Università degli Studi di Torino, Italy
2
Nanyang Technological University, Singapore
3
Technical University of Valencia, Spain
{bosco,patti,rossana}@di.unito.it, cambria@ntu.edu.sg, prosso@dsic.upv.es
The 2nd International Workshop on Emotion and Sentiment in Social and
Expressive Media (ESSEM 20151 ) is taking place on May 5, 2015, in Istanbul
as a workshop of the 14th International Conference on Autonomous Agents and
Multiagent Systems (AAMAS 2015). Emotions play a key role in the interactions
that occur in a multi-agent system. Relevant perspectives include, on the one
hand, research on architectures and cognitive models, which is concerned with
the integration of emotional states into agents and the role of emotions in agent
communication; on the other hand, research on techniques for sentiment analysis
and opinion mining, devoted to automatic processing of affective information
conveyed by spontaneous, multi-faceted user responses about shared contents.
The main goal of the workshop is to attain cross-fertilization between the two
perspectives. While the former relies mainly on cognitively inspired agent models,
the latter relies on the use of statistical and learning techniques supported by
resources such as corpora and linguistic datasets. By proposing ESSEM 2015
as an AAMAS workshop, we intend to stimulate a tighter integration of agent-
based paradigms with techniques for sentiment and opinion mining, which have
raised a growing interest in a social web and big data perspective. Since they
typically involve interaction and feedback as part of their functioning, social and
expressive media – ranging from online communities, blogs and fan-generated
narratives to interactive art installations – will provide an effective testbed for:
– integrating complementary aspects: emotion generation and detection, affect
expression and reception;
– studying the role of affect in the generation of social behavior, and commu-
nicative behavior in particular.
As the paradigmatic situation, we envisage a range of applicative contexts
that involve a performance of some kind, targeted to some audience/user. The
performance/reception paradigm, be it a digitally mediated co-creation session
or the contribution of a user in a social forum, provides a conceptual framework
for studying affect generation and detection in an integrated approach. Reception
can take the form of immediate feedback, possibly regulated by a protocol, or of
asynchronous response (e.g. tags and comments) conveyed through a plurality of
1
http://di.unito.it/essem15
media, against a background shaped by reputation and trust as a relevant part
of the audience value systems.
Given the leading thread described above, we have proposed a special focus
for ESSEM 2015: opportunities and challenges for emotion-aware multi agent
systems. The final program includes four full papers (out of the eight submis-
sions that were reviewed in the full paper category) and six short papers, which
all in all cover different topics of the call for papers and represent an inter-
esting variety of point of views on the ESSEM themes. The members of the
program committee did an exceptional job, managing to complete the review
process in record time, by providing authors with three reviews per full/short
paper, we would like to thank them for their accurate work. We thank our in-
vited speakers Ana Paiva (INESC-ID, Portugal) and Gülşen Eryiǧit (Istanbul
Technical University, Turkey) for accepting to deliver the ESSEM keynote talks
for this edition, crucial contributions towards a successful cross-fertilization of
ideas between the agent and NLP communities about emotion and sentiment
themes. We also thank Catherine Pelachaud (panel chair, CNRS, TELECOM
ParisTech, France), Chloé Clavel (Telecom-ParisTech, France), Emiliano Lorini
(IRIT-CNRS, Toulouse, France), Munindar P. Singh (North Carolina State Uni-
versity, USA), Gualtiero Volpe (University of Genova, Italy) which accepted to
animate our final discussion panel. We would like to express our gratitude for the
official endorsement we received from the CELI Torino, CIRMA and WIQ-EI
(Web Information Quality Evaluation Initiative). Furthermore, we are grateful
to all authors who submitted their works to ESSEM 2015.
We sincerely hope that ESSEM 2015 could be the occasion for merging the
the agent community and the sentiment analysis and opinion mining commu-
nity, both interested in emotions and sentiments in social and expressive media,
although from complementary perspectives. Still many are the open questions
for the two research communities. Does the progress in sentiment analysis and
opinion mining also bring opportunities for modeling the affective component
of agent behaviour? Are technologies for automatic sentiment analysis mature
enough to allow for a tighter integration with emotions in modelling feedback in
expressive media? Which affective frameworks should be considered in a perspec-
tive that integrates both the affective information extracted from user responses
and the affective agent behavior to be generated? Are there any relationships
between the analysis of emotion and sentiment in social and expressive media
and the captology research field, e.g. the study of computers as persuasive tech-
nologies? Hopefully, some of these questions will have a tentative answer in the
context of the ESSEM 2015 workshop.
April 2015 Cristina Bosco
Erik Cambria
Rossana Damiano
Viviana Patti
Paolo Rosso
Organization
Program Chairs and Organizers
Cristina Bosco University of Turin, Italy
Erik Cambria Nanyang Technological University, Singapore
Rossana Damiano University of Turin, Italy
Viviana Patti University of Turin, Italy
Paolo Rosso Universitat Politècnica de València, Spain
Program Committee
Alexandra Balahur European Commission Joint Research Centre, Italy
Cristina Battaglino University of Turin, Italy
Paula Carvalho INESC-ID and ISLA Campus Lisboa, Portugal
Marc Cavazza Teesside University, UK
Chloé Clavel Telecom-ParisTech, France
Morena Danieli University of Trento, Italy
Dipankar Das Jadavpur University, India
Mehdi Dastani Utrecht University, the Netherlands
Berardina Nadja De Carolis University of Bari Aldo Moro, Italy
Elisabetta Fersini University of Milano-Bicocca, Italy
Giancarlo Fortino University of Calabria, Italy
Carlos A. Iglesias Universidad Politécnica de Madrid, Spain
Anup Kalia North Carolina State University, Releigh, USA
Iolanda Leite Technical University of Lisbon, Portugal
Emiliano Lorini IRIT-CNRS, Toulouse, France
Viviana Mascardi University of Genova, Italy
Alessandro Moschitti University of Trento, Italy
Malvina Nissim University of Groningen, the Netherlands
Nicole Novielli University of Bari Aldo Moro, Italy
Antonio Origlia University of Naples Federico II, Italy
Roberto Paredes Technical University of Valéncia, Spain
Paolo Petta Austrian Research Institute for AI, Austria
Daniele Radicioni University of Turin, Italy
Francisco Rangel Autoritas Consulting, Spain
Hassan Saif The Open University, UK
Bjoern Schuller Technical University of Munich, Germany
Giovanni Semeraro University of Bari, Italy
Michael Thelwall University of Wolverhampton, UK
Marko Tkalcic Johannes Kepler University, Austria
Emilio Vivancos Technical University of Valéncia, Spain
Gualtiero Volpe University of Genova, Italy
Enrico Zovato Nuance Communications, Italy
Publicity Chair
Cristina Battaglino University of Turin, Italy
Endorsements
CIRMA, Università di Torino
CELI s.r.l Torino
WIQ-EI (Web Information Quality Evaluation Initiative)