Page 221 - DLIS002_KNOWLEDGE ORGANIZATION CLASSIFICATION AND CATALOGUING THEORY

Knowledge Organization: Classification and Cataloguing Theory

Notes 10. According to the implications of the general theory of the Subject Indexing Language, the
standard vocabulary that controls the use of the different concepts, has got to be a
………………………thesaurus.

11.3 Preserved Context Indexing System (PRECIS)

PRECIS (Preserved Context Index System) was designed and developed by Derek Austin by
about 1970 as an alternative procedure for deriving the subject headings and subject index
entries for British National Bibliography (BNB). Since 1952, for nearly 20 years, BNB followed
chain procedure for deriving subject index entries. Two most important factors worked for the
development of PRECIS: (i) Idea of replacing chain indexing technique of BNB; and (ii) the
decision of the British Library to generate computer produced BNB with all the indexes in view
of launching the UKMARC project. Accordingly, a research project for a suitable alternative for
generating subject indexes directly from the machine readable records were undertaken by the
British Library with the following objectives, which ultimately resulted in the development of
,
PRECIS:
(i) The computer, not the indexer, should produce all index entries. The indexer’s responsibility
would be only to prepare the input strings and to give necessary instructions to the
computer to generate indexes according to a definite format.
(ii) Each of the sought terms should find index entries and each entry should express the
complete thought content/full context of the document unlike the chain indexing where
only one entry is fully co-extensive with the subject and others are cross references
describing only one aspect of the complete content of the document.

(iii) Each of the entries should be expressive.
(iv) The, system should be based on a single set of logical rules to make it consistent.
(v) The system must have sufficient references for semantically related terms. PRECIS is now
a recognized indexing model and has been adopted by a number of indexing agencies to
produce subject indexes to a wide range of indexing tools like national bibliographies,
library catalogues, indexes to audio-visual materials, musical scores, micro documents,
etc.

The formation of subject headings in PRECIS is done in two stages. In the first stage, a human
indexer does all the intellectual tasks of subject analysis to set the indexing terms in an input
string according to the scheme of Role Operators. In the second stage, the computer generates
the desired index entries from the input strings prepared by the human indexer.
In carrying out these tasks, the indexer needs to keep the following factors in view:
(a) An entry can be made under any terms likely to be sought in a string;

(b) Each entry should be intelligible, and it should state the subject unambiguously; and
(c) Entries should be consistent in structure, so that they collocate with those produced from
other strings on similar themes.

The important criteria listed as (b) and (c) above can be met most effectively if the order of terms,
in an input string, and also in an index entry, is founded on the principle of “context dependency”.
In order to set down the selected indexing terms from a document in the sequence of context
dependency and to ensure that a team of indexers on different occasions, consistently arrive at
the, same conclusions concerning the preparation of input string, a scheme of “Role Operators”
are used. This `Scheme of Role Operators’ specifies the grammatical role or functions of the term
and regulates the order of terms in an input string, and thus establishes the relationships between

216 LOVELY PROFESSIONAL UNIVERSITY