|
Origins
The African Studies Thesaurus is the culmination of a two-pronged
project designed to transform the ASC Leiden library catalogue into a
comprehensive bibliographical tool in the field of African Studies and
facilitate total and easy national and international access to Africa
resources. The project was initiated in 1998 and in 1999 received
funding from NWO, the Netherlands Organization for Scientific Research.
The first prong, the conversion of the library’s card catalogue into
electronic form, involved the retrospective online cataloguing of some
75,000 titles and was completed in 2000.
The second prong, started in 1999, aimed to convert the numerical codes
used for subject access in the library’s online catalogue into a more
user-friendly word-based system. The codes, based on the UDC (Universal
Decimal Classification), had been derived from successive editions over
a period of almost 50 years, and had, generally, not been updated.
Moreover, the codes had also been adapted to suit the specific
requirements of an African Studies collection. An analysis of the
structure and consistency of these codes and an assessment of existing
language-based subject indexing systems used in other Africana libraries
led to the conclusion that there was no ready-made word-based system
which suited our needs. We therefore decided to build our own African
Studies Thesaurus.
Thesaurus construction
After an initial phase in which the technical and organizational aspects
of the project were charted and a thesaurus building tool was selected,
work on “translating” the UDC codes into descriptors and building a
thesaurus began in 2001. In the course of the following years modules
for African Languages (350 preferred terms), African Peoples (1200
preferred terms) and African Polities (300 preferred terms) were
developed for experimental use and placed on the ASC website. By March
2006 the last UDC codes had been “thesaur-ised”. More than 8,500 UDC
codes had been “translated” into some 5,200 descriptors (preferred
terms) and embedded in a thesaurus structure, with cross-references (c.
6,900 non-preferred terms), hierarchical and associative relationships .
The challenge of the project has been to combine two analytically
distinct activities inextricably linked in practice – the construction
of a thesaurus on the one hand, and the conversion of (UDC) codes into
descriptors on the other. The conceptual differences between a numerical
code in a classification and words in a natural language were brought
home with force. The task was further complicated by inconsistent
classification practice over the years and shifts in the meaning and
coverage of codes. There was often no one-to-one match between a code
and a thesaurus descriptor and a good many previously assigned UDC codes
had to be “cleaned up”. Because of the many different kinds of relations
between codes and descriptors, the automatic conversion of all the UDC
codes assigned to titles in the ASC catalogue into descriptors was a
technically complex process.
Implementation
An ongoing concern throughout the project was the implementation and
operationalization of the thesaurus in cooperation with OCLC PICA, which
hosts the library’s online catalogue. For a number of reasons it was
decided not to integrate the thesaurus fully into the catalogue but to
make it available on the ASC website instead, with reciprocal links
between the catalogue and the thesaurus. In April 2006 the first steps
to operationalize the thesaurus were taken. This was followed by the
conversion of all the UDC codes into descriptors and the incorporation
of descriptors in the catalogue. With the publication of the African
Studies Thesaurus on the ASC website in June 2006 the project was
completed. |
|
Send your comments to
Katrien Polman |