[Corpora-List] Call for Papers: Stylistics in Text Retrieval Practice

From: Ozlem Uzuner (ozlem@mit.edu)
Date: Wed May 03 2006 - 19:40:55 MET DST

  • Next message: TadPiotr: "RE: [Corpora-List] Bible corpora?"

                  Stylistics for Text Retrieval in Practice
                             SIGIR 2006 Workshop
                           Seattle, August 10, 2006

                     http://www.lingcog.iit.edu/style2006
                              style2006@sics.se

      Recent years have seen an increased attention to various aspects
      of automatic analysis and extraction of stylistic aspects of
      natural language texts.

      Style may be roughly defined as the 'manner' in which something is
      expressed, as opposed to the 'content' of a message. Modelling,
      representing, explaining, and utilizing variation in the manner of
      expression is the business of stylistic analysis.

      This workshop follows four previous successful exploratory events
      on stylistic analysis and will focus on the practical craft of
      stylistic analysis in natural language texts.

                          THIS YEAR: BRING A DEMO!

      Potentially useful applications of stylistic analysis abound,
      including systems for genre-based information retrieval,
      authorship attribution, plagiarism detection, context-sensitive
      text or speech generation systems, organizing and retrieving
      documents based on their writing style, attitude, or sentiment,
      quality or appropriateness filters for messaging systems,
      detecting abusive or threatening language, and more. This year,
      participants are expected to bring with them a method for applying
      stylistic analysis to information access tasks.

      Before lunch, methods are discussed in session; after lunch,
      demonstrated in practice.

                            DISCUSSION QUESTION

      Participants should address the following key challenge question
      in their participation proposals:

      * WHAT IS A MEANINGFUL "KILLER APP" FOR STYLISTIC TEXT ANALYSIS? *

      and consider the following questions for discussion in session:

      1. How does style relate to other forms of non-topical textual variation?
      2. What features are best for different style analysis tasks?
      3. Is cross-lingual or 'universal' style analysis possible, and if so, how?
      4. How might we develop useful shared resources for moving style research forward?

                             SUBMISSION FORMAT

      Send us a statement (in PDF) of up to five pages describing your
      research or application with a short description (and screenshots
      if possible) of the demonstration you plan to show. At the
      workshop we will discuss and decide on an appropriate forum for a
      more permanent record of the proceedings. If you would like a
      speaking slot in the discussion session, you should indicate this
      in your statement of interest.

                                ORGANIZERS

      Shlomo Argamon, Illinois Institute of Technology, USA
      Jussi Karlgren, Swedish Institute for Computer Science, Sweden
      Ozlem Uzuner, University at Albany, State University of New York, USA

                     http://www.lingcog.iit.edu/style2006
                              style2006@sics.se

                              IMPORTANT DATES

      Expression of interest to participate now!
      Submission of participation proposals June 1
      Notification of acceptance June 14
      Program published June 20
      Workshop August 10



    This archive was generated by hypermail 2b29 : Wed May 03 2006 - 20:31:46 MET DST