MPQA: Multi-Perspective Question Answering


Leader: Prof. Janyce Wiebe

Participants: Eric Breck, Chris Buckley, Claire Cardie, Paul Davis, Bruce Fraser, Diane Litman, David Pierce, Ellen Riloff, Theresa Wilson

Products:

The new version of the corpus, MPQA Opinion Corpus version 1.1 is available for download

In version 1.1, the annotation terminology has been updated to match the new and clearer terminology used in
(Wiebe, Wilson and Cardie, Annotating expressions of opinions and emotions in language, Language Resources and Evaluation 1(2), 2005). A link to this paper is at http://www.cs.pitt.edu/~wiebe/pubs/pub1.html.

In addition, incorrect sentence splits have been manually corrected,along with any annotations that were affected.

Browse or download the final briefing developed at the conclusion of the workshop.
View a final report on the work carried out in this workshop. (Download postscript version of report.)
View the annotation guidelines developed and used within the course of the workshop.

View an example news report of the kind annotated within this workshop. View an image of how the data was annotated using the GATE annotation tool. View an image depicting the high-level opinion annotations added to the example news report. View an image depicting the low-level opinion annotations added to the example news report.

Overview:
A group of resarchers and PhD students worked together to explore the area of Multi-Perspective Question Answering (MPQA). The accomplishments include a knowledge representation scheme to support manual annotation and analysis of data; a repository of linguistic clues relevant for perspective; a data corpus; a set of manually annotated data; an annotation system to support manual annotation; an application architecture; and the results of various types of evaluation.

The problem we addressed is finding and organizing expressions of opinions in the world press and other text. Our work builds toward the following tasks to support activities of professional information analysts.

  • Given a particular topic, event, or issue, find a range of opinions being expressed about it in the world press.
  • Once opinions have been found, clustering them and their sources in various ways. The {\it source} of an opinion or perspective is simply the person or group whose opinion or perspective it is. There are various attributes according to which opinions and their sources may be clustered, including:
  • The type of attitude that is expressed. For example, the source might be expressing a positive, negative, or uncertain attitude.
  • The basis for the opinion, such as supporting beliefs, or experiences
  • The expressive style of the sentences. The style might be sarcastic and vehement, for example, or neutral.
  • Once systems are developed to automate the above tasks, they may be applied to many topics and documents, to build perspective profiles of various groups and sources, and observe how attitudes change over time.

    •  

    If you are interested in obtaining any of the data generated in the course of this workshop, please contact David Day, day@mitre.org.