THE DESIGN OF BROWSING AND BERRYPICKING TECHNIQUES

FOR THE ONLINE SEARCH INTERFACE (1989)by Marcia J. Bates

Information Retrieval

February 24, 2004

Summary

Berrypicking -- a new model of searching that is closer to the real behavior of information searchers than the traditional model of information retrieval. Based on manual environment, but used to inform electronic one.

Literature Review to find interface capabilities that are desired by users.

Suggestions for search engines and interfaces in the berrypicking context.

Traditional Model of IR

Fundamental to the traditional model of IR is the idea of a single query presented by the user, matched to the database contents, yielding a single output set.

Query is treated as a single unitary, one-time conception of the problem.

Salton Modification to Traditional IR Model

Iterative feedback to improve output.

Salton developed a system that would modify the query formulation based on user feedback to the first preliminary output set.

Only applies to query formulation. For Salton, information need or problem is treated as unchanging.

This is still a traditional IR model.

Formalized IR Process (Gheorghe)

Not the same as traditional model

Both query and information need can evolve and be reformulated depending on results of search

Very close to Berrypicking Model

Real Life Upends the Traditional IR Theory

Traditional theory does not describe real users with real information needs.

In real-life, users begin with one feature of a broader topic and move through various sources.

Each new piece of information gives the user new ideas and directions and thus a new conception of his/her query.

This is not simply a change in search terms. Rather the query or information need itself (and the search terms used) is continually evolving.

At each stage the user identifies useful information and references.  In other words, the query is satisfied not by a single final retrieved set, but by a series of selections of individual references and bits of information at each stage of the ever-modifying search.  A bit-at-a-time retrieval of this sort is called berrypicking.

A Berrypicking Model of IR

Compare Traditional and Berrypicking

Nature of the Query

Nature of the Overall Search Process

Range of Search Techniques Used

Information “Domain" or Territory where the Search is Conducted

Nature of the Query

Evolving rather than single and unchanging.

Nature of the Search Process

Follows a berrypicking pattern rather than a straight line leading to a single best retrieved set.

Range of Search Techniques Used

Actual ways users search in manual environments. Footnote chasing Citation searching Journal run Area scanning Subject searches in bibliographies and

abstracting and indexing (A & I) services Author searching

Range of Search Techniques Used

Automated systems focus only on searching abstracting and indexing services. [Dated supposition].

Real life involves all of the noted techniques used in endless variation.

“From the standpoint of general effectiveness in searching, it is clear, on reflection, however, that, other things being equal, the searcher with the widest range of search strategies available is the searcher with the greatest retrieval power.”

Solution – Incorporate all techniques into future electronic search interfaces.

Information “Domain" or Sources where the Search is Conducted

Varied and continually changing for any given query/information need.

Otherwise, not expanded/explained by Bates.

Suggestions for Berrypicking Search Interface (Importance of Browsing)

In many ways, berrypicking is like browsing.

Because of the importance of browsing in information retrieval, sophisticated browsing techniques should be built into an interface to enhance its berrypicking capabilities.

Suggestions for Berrypicking Search Interface (Technique Specific)

Footnote chasing Ability to browse through the article or book that

generates the references as well as through the list of references--in fact, to move back and forth easily between the two parts of the document.

Design Suggestion -- User can get the following easily, preferably by direct manipulation, e.g., with mouse and pull-down menus: 1) Overview of document contents--chapter or section headings, 2) Full text of documents and references, 3) Ability to jump back and forth between text and references.

Suggestions for Berrypicking Search Interface (Technique Specific)

Citation searching Ability to browse the set of references that cite a given

starter reference, or read any of the citing articles. Design Suggestion -- Users should have the ability to 1)

Scan lists of citing references, 2) Make simple single step jumps to a) full text of citing articles, b) full list of references in citing article,  and 3) Make jumps in any direction ad infinitum, i.e., the user should not have to "return to go" and reenter a starting article for each jump in any direction.

Suggestions for Berrypicking Search Interface (Technique Specific)

Journal run Ability to review the contents lists or articles in a

journal. Design suggestion -- 1) Easy specification of

journal title and starting date in a journal run search, 2) Easy jumps between contents lists and articles and back again, 3) Capability of requesting, if wanted, standard section headings in scholarly articles, such as "Methodology," or "Conclusions," so the searcher is shown these sections directly.

Suggestions for Berrypicking Search Interface (Technique Specific)

Area scanning Ability to browse books or materials by subject matter

whether on the library’s shelves or by some other method.

Design suggestion -- 1) A library's listing of its books on the shelves arranged by the order of the classification scheme is called a shelf list.  Thus, for area scanning linearly along the shelves, a capability of browsing the shelf list can be provided.  2) For "jumping the rails" of the classification scheme, browsing at several levels of generality within the classification scheme itself can be provided.  3) At any point, with either of the first two capabilities listed in this section, the searcher should be able to ask for "snapshots" of full text of books.

Suggestions for Berrypicking Search Interface (Technique Specific)

Subject searching in bibliographies and A & I services Ability to browse identifiers or descriptors in

database. Design suggestion -- The user should have the

capability of 1) Rapid browsing of many references without cost, and/or ability to ask to see every nth reference in a large set.  2)  Browsing the classification used in an A & I service, as well as abstracts within each classification, either all or every nth one.

Suggestions for Berrypicking Search Interface (Technique Specific)

Author searching Ability to browse different works by the same

author. Design suggestion -- When author searching,

the user should have the capability of calling up 1) Bibliographies of authors' works, 2) "Snapshots" of the text of works, and 3) Features that enable footnote chasing and citation searching.

Suggestions for Berrypicking Search Interface (Overarching)

Large flexible databases Virtual layout of physical library Ability to move rapidly through text (aspect of

randomness) Ability to see random samples of books or other

materials quickly Noted search techniques should be accessible easily Hypertext High definition screens for easy reading and scanning

(need to view large amounts of information at once) Easy to highlight or mark for storage desired references

or citations.

Berrypicking Model Supported by Literature

LineHogeweg-de Haart StoneStoanEllisKuhlthau

Conclusion

Article shows its age.Prescient. Most of suggestions have been

implemented.Intuitively, seems to be an accurate

description of how users seek information.But --- No Data.Is this theory?