Structuring collections with Scatter/Gather extensions

ABSTRACT

A major component of sense-making is organizing--grouping, labeling, and summarizing--the data at hand in order to form a useful mental model, a necessary precursor to identifying missing information and to reasoning about the data. Previous work has shown the Scatter/Gather model to be useful in exploratory activities that occur when users encounter unknown document collections. However, the topic structure communicated by Scatter/Gather is closely tied to the behavior of the underlying clustering algorithm; this structure may not reflect the mental model most applicable to the information need. In this paper we describe the initial design of a mixed-initiative information structuring tool that leverages aspects of the well-studied Scatter/Gather model but permits the user to impose their own desired structure when necessary.

REFERENCES

• 1: Douglass R. Cutting , David R. Karger , Jan O. Pedersen , John W. Tukey, Scatter/Gather: a cluster-based approach to browsing large document collections, Proceedings of the 15th annual international ACM SIGIR conference on Research and development in information retrieval, p.318-329, June 21-24, 1992, Copenhagen, Denmark [doi>10.1145/133160.133214]
• 2: P. Pirolli. Information Foraging Theory. Oxford University Press (2007).
• 3: Peter Pirolli , Patricia Schank , Marti Hearst , Christine Diehl, Scatter/gather browsing communicates the topic structure of a very large text collection, Proceedings of the SIGCHI conference on Human factors in computing systems: common ground, p.213-220, April 13-18, 1996, Vancouver, British Columbia, Canada [doi>10.1145/238386.238489]
• 4: R. Mihalcea and P. Tarau, "TextRank: Bringing Order into Texts", Proc. of EMNLP (2004).
• 5: M. Hearst, D. Karger, and J. Pedersen. "Scatter/Gather as a Tool for the Navigation of Retrieval Results", Proc. Of AAAI (1995).

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.3 Information Search and Retrieval
          Subjects: Clustering

Additional Classification:
  H. Information Systems
  H.5 INFORMATION INTERFACES AND PRESENTATION (I.7)
      H.5.2 User Interfaces (D.2.2, H.1.2, I.3.6)
          Subjects: Prototyping

General Terms:
Design, Experimentation

Keywords:
Scatter/Gather, clustering, sense-making, text summarization