IEEE Intelligent Special Issue: Concept-Level Opinion and Sentiment Analysis
Catherine Havasi
http://www.computer.org/portal/web/computingnow/iscfp2
Submission deadline: 1 July 2012
Opinions play a primary role in decision-making processes. Whenever
people need to make a choice, they are naturally inclined to hear
others’ opinions. In particular, when the decision involves consuming
valuable resources, such as time and/or money, people strongly rely on
their peers’ past experiences. Just a few years ago, the main sources
for collecting such information were friends, acquaintances and, in
some cases, specialized magazines or websites.
The passage from a read-only to a read-write Web has provided people
with new tools that allow them to create and share, in a timely and
cost-efficient way, their own contents, ideas, and opinions with
virtually millions of people connected to the World Wide Web. The
opportunity to capture the opinions of the general public about social
events, political movements, company strategies, marketing campaigns,
and product preferences has raised more and more interest both in the
scientific community, for the exciting emergent challenges, and in the
business world, for the remarkable fallouts in marketing and financial
market prediction.
Mining opinions and sentiments from natural language, however, is an
extremely difficult task: it involves a deep understanding of most of
the explicit and implicit, regular and irregular, syntactical and
semantic rules of a language. Existing approaches mainly rely on parts
of text in which opinions and sentiments are explicitly expressed such
as polarity terms, affect words, and their co-occurrence frequencies.
However, opinions and sentiments are often conveyed implicitly through
latent semantics, which make purely syntactical approaches
ineffective.
In this light, this special issue focuses on the introduction,
presentation, and discussion of novel approaches to opinion mining and
sentiment analysis that are not entirely based on domain-dependent
corpora but also on general-purpose semantic knowledge bases. The main
motivation for the issue, in particular, is to go beyond a mere
word-level analysis of text and provide novel concept-level approaches
to opinion mining and sentiment analysis that allow a more efficient
passage from (unstructured) textual information to (structured)
machine-processible data, in potentially any domain. Articles are thus
invited in areas such as AI, the Semantic Web, knowledge-based
systems, and adaptive and transfer learning for research on opinion
and sentiment retrieval and analysis. Potential topics include:
* Opinion and sentiment summarization and visualization
* Explicit and latent semantic analysis for opinion and sentiment mining
* Knowledge base construction and integration with opinion and
sentiment analysis
* Transfer learning of opinion and sentiment with knowledge bases
* Time-evolving opinion and sentiment analysis
* Corpora and resources for opinion and sentiment analysis
* Multimodal sentiment analysis
* Multidomain and cross-domain evaluation
* Multilingual sentiment analysis and reuse of knowledge bases
Guest Editors:
• Erik Cambria, National University of Singapore, Singapore
• Bjoern Schuller, Technische Universitat Munchen, Germany
• Bing Liu, University of Illinois at Chicago, USA
• Haixun Wang, Microsoft Research Asia, China
• Catherine Havasi, MIT Media Laboratory, USA
Submission Guidelines:
The special issue will consist of papers on novel methods and
techniques for building and using semantic knowledge bases in the
field of opinion mining and sentiment analysis. Besides the specified
topics of interest, the special issue also welcomes papers on specific
application domains of sentiment analysis—for example, social data
mining, influence networks, customer experience management,
computer-mediated human-human communication, social media marketing,
multimedia management, personalization and persuasion, enterprise
feedback management, human-agent, -computer and -robot interaction,
intelligent user interfaces, patient opinion mining, surveillance, and
art.
Submissions should be 3,000 to 5,400 words (counting a standard figure
or table as 200 words) and should follow IEEE Intelligent Systems
style and presentation guidelines
(www.computer.org/intelligent/author). The manuscripts cannot have
been published or be currently submitted for publication elsewhere. We
strongly encourage submissions that include audio, video, and
community content, which will be featured on the IEEE Computer Society
Web site along with the accepted papers.
Questions?
• For general information about the special issue, contact Erik
Cambria (include the keyword “concept-level sentiment analysis” in the
subject line) at cam...@nus.edu.sg.
• For general author guidelines, see www.computer.org/intelligent/author.
• For submission details, see intel...@computer.org.
• To submit an article, go to
https://mc.manuscriptcentral.com/is-cs (log in and then select
“Special Issue on Concept-Level Sentiment Analysis”).