Contact: wassa2023@googlegroups.com Website: https://wassa-workshop.github.io/

BACKGROUND AND ENVISAGED SCOPE

Subjectivity and Sentiment Analysis has become a highly developed research area, ranging from binary classification of reviews to the detection of complex emotion structures between entities found in text. This field has expanded both on a practical level, finding numerous successful applications in business, as well as on a theoretical level, allowing researchers to explore more complex research questions related to affective computing. Its continuing importance is also shown by the interest it generates in other disciplines such as Economics, Sociology, Psychology, Marketing, Crisis Management & Digital Humanities.

The aim of WASSA 2023 is to bring together researchers working on Subjectivity, Sentiment Analysis, Emotion Detection and Classification and their applications to other NLP or real-world tasks (e.g. public health messaging, fake news, media impact analysis, social media mining, computational literary studies) and researchers working on interdisciplinary aspects of affect computation from text. For this edition, we encourage the submission of long and short research and demo papers including, but not restricted to the following topics:

• Resources for subjectivity, sentiment, emotion and social media analysis • Opinion retrieval, extraction, categorization, aggregation and summarization • Humor, Irony and Sarcasm detection • Mis- and disinformation analysis and the role of affective attributes • Aspect and topic-based sentiment and emotion analysis • Analysis of stable traits of social media users, incl. personality analysis and profiling • Transfer learning for domain, language and genre portability of sentiment analysis • Modelling commonsense knowledge for subjectivity, sentiment or emotion analysis • Improvement of NLP tasks using subjectivity and/or sentiment analysis • Intrinsic and extrinsic evaluation of subjectivity and/or sentiment analysis • The role of emotions in argument mining • Application of theories from related fields to subjectivity and sentiment analysis • Multimodal emotion detection and classification • Applications of sentiment and emotion mining • Public sentiments and communication patterns of public health emergencies.

We furthermore encourage submissions to the special theme Ethics in Affective Computing, including opinion papers, as well as experimental papers. This includes the following topics, but is not limited to them:

• Which properties of a model render a automatic analysis task unethical? • Which characteristics of an annotation task are to be considered in ethical considerations? • What are appropriate methods to analyze data and models from an ethical perspective? • What aspects are particular important for affective analysis tasks, in contrast to other NLP settings?

IMPORTANT DATES April 24, 2023 – Submission deadline for main workshop papers. May 1, 2023 – Commitment deadline for submitting through ARR with reviews May 22, 2023 – Notification of acceptance. June 6, 2023 – Camera-ready papers due. June 12, 2023 – Pre-recorded video due. July 13 or 14, 2023 – Workshop.

Note that the shared tasks follow a different timeline that will be communicated separately.

SUBMISSION At WASSA 2023, we will accept four types of submissions: long, short, ARR commitments, and industry track demo papers. For the regular research track we accept long & short papers. Submission is electronic, through the OpenReview portal for the workshop with the deadline on April 24, 2023. Both long and short papers must be anonymised for double-blind reviewing, must follow the ACL Author Guidelines, and must use the ACL 2023 templates available on the ACL Rolling Review website. The submitting author must have an OpenReview profile. Please ensure profiles are complete before submission.

Long: Long papers may consist of up to eight (8) pages of content, with any number of additional pages of references, and will be presented orally.

Short: Short papers may consist of up to four (4) pages of content, with two (2) additional pages of references, and will be presented either orally or as a poster.

ARR Commitments: Additionally, we accept double submissions and double commitment of ARR reviews in parallel to WASSA and another venue. Please note that you must immediately withdraw your paper from WASSA if you decide to publish it elsewhere. They must be committed to the workshop (together with the reviews) not later than May 1, 2023.

Industry Demos: We also include an industry track, for which we accept demo papers that describe system demonstrations, ranging from early prototypes to mature production-ready systems. Please note: Commercial sales and marketing activities are not appropriate for this track. Demo papers may consist of up to six (6) pages of content, these will be presented as a poster and should include a live demonstration.

Additionally, system description papers from the shared tasks will be presented either orally or as poster.

SHARED TASK Following the success of the shared tasks organized in 2017, 2018, 2021 and 2022, we will continue our line of shared tasks. We will propose a first shared task on Empathy Detection and Emotion Classification in conversation at the speech-turn level, and a second shared task on multi-class and multi-label emotion classification on code-mixed (Roman Urdu + English) text messages. The tasks and deadlines will be communicated in due time. Keep a close eye on the workshop website for more details: https://wassa-workshop.github.io/

ORGANIZERS Jeremy Barnes, IXA group, University of the Basque Country UPV/EHU Orph ́ee De Clercq, LT3 Language and Translation Technology Team, Ghent University Roman Klinger, Institut f ̈ur Maschinelle Sprachverarbeitung, University of Stuttgart, Germany Valentin Barriere, Centro Nacional de Inteligencia Artificial Shabnam Tafreshi, University of Maryland: ARLIS Jo ̃ao Sedoc, Technology, Operations, and Statistics department, New York University Iqra Ameer, Yale University