Page 247 - DLIS402_INFORMATION_ANALYSIS_AND_REPACKAGING

Information Analysis and Repackaging

Notes — Assist the development of search strategies by a user. The user can browse through the the-
saurus in search of the most appropriate terms for his/her particular query.
— Refine a query, either by
• Reformulating it with broader terms (query expansion), useful when a query has returned
too few relevant results.
• Query contraction, by reformulation with narrower terms
It is currently used to index manuals (e.g., corporate policy and procedure manuals), large contracts
and large quantities of e-mail. Technical writers who index their own work have been using it as a
first step in indexing.
Among indexers, Indexicon is most likely to be useful for specialists, who are more likely to take
the time to create specialised lexicons, and to work with the program to enhance its efficacy in their
special field. For journal indexing, where the same indexer works with similar material, in a consistent
format, year after year, it might be worth taking the trouble to set up a specialised lexicon, and use
Indexicon as a first step. But Indexicon is not good enough at picking key concepts and leaving out
worthless ones, to be useful, in general, as an aid to indexing books.
If Indexicon improves, and if the embedded indexing software used in word processing programs
improves, it may become more cost-effective to start indexing with Indexicon, and then enhance
the index by editing.
As the ability of computer software to recognise personal names develops, it may also become
useful as a tool for automatically generating name indexes (Feldman, Lawrence e-mail 15/03/96).

Effect of Automatic Methods on Professionals
As computer programs become more sophisticated, and more information appears in electronic
form, there will eventually be less ‘traditional’ indexing work available. This loss may be balanced
in the short-term by an increase in the number of databases and an increase in the number of indexing
and abstracting projects attempted. The proportion of freelance versus in-house work may also
change.
Humans should still be used for important works, which perhaps can be identified by studying
usage and citation patterns (Anderson 1993). Indexers and abstracters will have to become more
selective, and decide on the quality of the works they might index and abstract, as well as the
subject content.
If we remain better than computers we must show this, and indicate that there are economic returns
(to the publisher) and academic returns (to the index or abstract user) from a quality index or abstract.
On the positive side, indexing and abstracting skills will be needed in the development of computer
systems, and to check the output from computers. Indexers will be needed to set up and maintain
thesauruses, and to train writers as ‘bottom-up indexers’ so that their work is readily retrievable.

Indexers will have to become entrepreneurial and computer literate. Indexers with
skills in the related areas of computing, editing, librarianship and bibliography
may be best suited to take advantage of new opportunities. We will have to be
able to identify gaps in the organisation of knowledge and to fill those gaps in a
commercially effective way. To do this we will have to be computer literate. Not
only will we have to know how to use various computer tools for indexing; we
will also have to know how information is organised and used electronically, so
that we can best understand the needs and make our own contributions.

Application of Thesauri to Information Retrieval (IR) Systems
In this document we will try to describe the application of thesauri to information retrieval (IR)
systems, underlining the differences between manual thesauri and automatically generated ones.

242 LOVELY PROFESSIONAL UNIVERSITY