

Human language technology systems have typically focused on the "factual" aspect of content analysis. Other aspects, including pragmatics, point of view, and style, have received much less attention. However, to achieve an adequate understanding of a text, these aspects cannot be ignored.
In this symposium, we address computer-based analysis of “point of view”. Our goal is to bring together people from academia, government, and industry to explore annotation, modeling, mining, and classification of opinion, subjectivity, attitude, and affect in text, across a range of text management applications.
The symposium therefore addresses a rather wide range of issues, from theoretical questions and models, through annotation standards and methods, to algorithms for recognizing, clustering, characterizing, and displaying attitudes and affect in text. Despite growing interest in this area, with papers recently published in major conferences and new corpora developed, there has never been a workshop or symposium that targets a wide audience of researchers and practitioners on these topics.
We expect focused discussions of current challenges, existing models, and future directions. A joint session with the "Architectures for Modeling Emotion: Cross-Disciplinary Foundations" symposium is planned.
We invite contributions on methodological, technical, and application-oriented aspects of this emerging subfield in text processing, including but not limited to the following list of topics.
Types and models of subjective information
Opinion, sentiment, point of view
Affect, emotion
Uncertainty, doubt, and related epistemic qualities
Annotation
Categorization, and characteristics such as centrality, polarity, intensity
Annotation at different levels of granularity (expression, clause, sentence, discourse segment, document, multi-document)
Inter annotator agreement studies
Tools for annotation
Bootstrapping using machine-learned classifiers
Semantic lexicons
Lists of affect-bearing words and phrases
Ontologies
Classification models
Identifying subjective/opinionated/affective expressions
Contextual disambiguation of potentially subjective expressions
Discourse segmentation
Clustering techniques
Summarization techniques
Fusion of points of view
Clustering techniques
Visualization tools
As a component technology
As a standalone technology
Government applications
Opinion and affect oriented question answering systems
Affect and opinion oriented retrieval and extraction systems
CRM (Customer Relation Management) system
Newsgroups and other text
Abstracts and full papers October 3, 2003
Notification of acceptance November 7, 2003
Final versions of abstracts and papers January 30, 2004
Application for Student Funding January 25, 2004
Symposium March 22 - 24, 2004
We have a limited amount of money to support graduate student travel. If you want to be considered for funding, please send an informal application (click here for student travel funding details) to the Symposium co-chairs by January 25, 2004
Submissions can be extended abstracts (three pages) or full papers (up to eight pages). Accepted papers will be published in the symposium proceedings.
Yan Qu,
(Co-Chair), Clairvoyance Corporation (yqu at clairvoyancecorp dot com)
James G.
Shanahan, (Co-Chair), Clairvoyance Corporation (jimi at clairvoyancecorp
dot com)
Janyce Wiebe, (Co-Chair), University of Pittsburgh (wiebe at cs dot pitt
dot edu)
Claire Cardie, Cornell University (cardie at cs.cornell dot edu)
Eduard Hovy, USC/Information Sciences Institute (hovy at isi dot edu)
Elizabeth Liddy, Syracuse University (liddy at mailbox dot syr dot edu)
| Clairvoyance talks | |
| Clairvoyance research | |
Last update of this page: February 18, 2004