A workshop collocated with:
|Call for papers||Important dates||Submission||Program Committee||Workshop organisation||Workshop organizers|
Call for papersThis workshop will investigate the relation and complementarity between factual and opinionated information for decision-making. Traditionally, natural language processing approaches and tools have segregated the objective and subjective treatments of information in two separate fields of expertise where the two never met. This workshop is an attempt to fill this gap and examine the relation and complementarity that exists between the factual and the speculative in order to build a more subtle representation of the state of affairs.
In their everyday routine, managers, professionals, even laymen, must deal with a large amount of textual information to help make their decisions. Without appropriate language technology tools to extract, classify and synthesize this information, decision-makers would have to read a large number of so-called objective sources but arguably even more subjective and opinionated material. Facts and opinions would then be combined together to create an accurate picture of the entity or issue we seek to understand. Facts are of course crucial to make professional or personal decisions, but opinions may sometimes play a key role. For this workshop we have identified two specific domains where factual and subjective information have an importance: brand management and health (patient empowerment and education), with an eye towards clinical decision. However, any relevant domains where the factual and the opinionated co-exist would be considered.
We encourage submissions addressing the following questions:
- Relevance of the extracted information (certainty, speculation and subjectivity) in a decision making context
- How to combine objective and subjective information in NLP-based decision making to produce a better understanding
- Exploitation of the objective and subjective information:
- How do managers combine the factual and the less tangible to favour the success of their brand?
- How does the patient understand a medical decision given objective information by the physician and subjective information related to his/her feelings, as well as discussions with other patients?
- How does a physician combine what the data show about a patient with more subjective information like symptoms or informed opinion from a third party?
We would like to invite anyone currently researching in the areas of sentiment analysis, information extraction, text mining, BioNLP or automatic summarization to submit a paper, including submissions presenting a discussion of a system.
October 1st: notification of acceptance
October 15th: camera-ready papers for publication
SubmissionPapers should be a maximum of 8 pages, all included and follow the SLTC format. Only pdf files are accepted.
Style files: The style files will be the same as for SLTC 2010:
LaTeX: sltc2010.tex sltc2010.sty sltc2010.bst sltc2010.pdf sltc2010.zip
MS Word: sltc2010.doc
Program committee members
- Khurshid Ahmad (Trinity College Dublin, Ireland)
- Dina Demner-Fushman (NIH/NLM/LHC, USA)
- Roger Evans (NLTG, University of Brighton, UK)
- Agata Jackiewicz (Université Paris-Sorbonne, France)
- Jonathon Read (University of Oslo, Norway)
- Ricardo Ribeiro (INESC-ID/ISCTE-IUL, Portugal)
- Harald Trost (CeMSIIS, Medical University of Vienna, Austria)
- Xinglong Wang (Brandwatch, UK)
- Pierre Zweigenbaum (LIMSI, France)
Workshop organisationThe duration of the workshop is three hours. There will be regular (short) talks and according to the number and the quality of the submissions, posters.
Every paper will be reviewed by 2 members of the program committee. All accepted papers will be published in the NEALT Proceeding Series. Authors of the accepted papers will have to sign an agreement with NEALT publishers.
Details on place and time are provided at the conference website.