Proceedings and Book of Abstracts

Workshop Programme

9:00 – 9.10 Introduction by Workshop Chair
9.10 – 10:30 Social Media
Chair: J. Fernando Sánchez
Cristina Bosco et al. Tweeting in the Debate about Catalan Elections
Ian D. Wood and Sebastian Ruder Emoji as Emotion Tags for Tweets
Antoni Sobkowicz and Wojciech Stokowiec Steam Review Dataset - new, large scale sentiment dataset
10:30 – 11:00 Coffee break
11:00 – 13:00 Corpora and Data Collection
Chair: Viviana Patti
Ebuka Ibeke et al. A Curated Corpus for Sentiment-Topic Analysis
Jasy Liew Suet Yan and Howard R. Turtle EmoCues-28: Extracting Words from Emotion Cues for a Fine-grained Emotion Lexicon
Lea Canales et al. A Bootstrapping Technique to Annotate Emotional Corpora Automatically
Francis Bond et al. A Multilingual Sentiment Corpus for Chinese, English and Japanese
13:00 – 14:00 Lunch break
14:00 – 15:00 Personality and User Modelling
Chair: Björn Schuller
Shivani Poddar et al. PACMAN: Psycho and Computational Framework of an Individual (Man)
Veronika Vincze1, Klára Hegedűs, Gábor Berend and Richárd Farkas Telltale Trips: Personality Traits in Travel Blogs
15:00 – 16:00 Linked Data and Semantics
Chair: Ian D. Wood
Minsu Ko Semantic Classification and Weight Matrices Derived from the Creation of Emotional Word Dictionary for Semantic Computing
J. Fernando Sánchez-Rada et al. Towards a Common Linked Data Model for Sentiment and Emotion Analysis
16:00 – 16:30 Coffee break
16:30 – 18:00 Beyond Text Analysis
Chair: J. Fernando Sánchez
Bin Dong, Zixing Zhang and Björn Schuller Empirical Mode Decomposition: A Data-Enrichment Perspective on Speech Emotion Recognition
Rebekah Wegener, Christian Kohlschein, Sabina Jeschke and Björn Schuller Automatic Detection of Textual Triggers of Reader Emotion in Short Stories
Andrew Moore, Paul Rayson and Steven Young Domain Adaptation using Stock Market Prices to Refine Sentiment Dictionaries

Scope

The Emotion and Sentiment Analysis workshop is the sixth edition of the highly successful series of Corpora for Research on Emotion. As its predecessors, the aim of this workshop is to  connect the related fields around sentiment, emotion and social signals, exploring the state of the art in applications and resources. All this, with a special interest on multidisciplinarity, multilingualism and multimodality.

This year's edition will again also put an emphasis on common models and formats, as a standardization process would foster the creation of interoperable resources. In particular, researchers are encouraged to share their experience with Linked Data representation of emotions and sentiment and present any initiatives to go beyond the state of the art. Beyond data modelling, the workshop will deal with harnessing Linked Open Data in every application: from data collection and sharing to data analysis.

Approaches on semi-automated and collaborative labeling of large data archives will also be of interest, such as by efficient combinations of active learning and crowdsourcing. In particular also for combined annotations of emotion, sentiment, and social signals. Multi- and cross-corpus studies (transfer learning, standardisation, corpus quality assessment, etc.) are further highly relevant, given their importance in order to test the generalisation power of models.

Topics include, but are not limited to:

Important Dates

Submission

Submitted abstracts of papers for oral and poster must consist of about 1500 – 2000 words. All submissions must follow the guidelines at LREC 2016. Camera-ready versions will be limited to:

All submissions will be done via the START page. When submitting a paper, authors will be asked to provide essential information about resources (in a broad sense, i.e. also technologies, standards, evaluation kits, etc.) that have been used for the work described in the paper or are a new result of your research.

Moreover, ELRA encourages all LREC authors to share the described LRs (data, tools, services, etc.) to enable their reuse and replicability of experiments (including evaluation ones)

Organising Committee

Program Committee

Sponsored by: