Thesaurus Construction and Use: A Practical Manual

Thesaurus Construction and Use: A Practical Manual

Thesaurus Construction and Use: A Practical Manual

Thesaurus Construction and Use: A Practical Manual

Synopsis

Written by leading experts in the field, this practical guide to the construction of thesauri for use in information retrieval includes topics such as planning and design; vocabulary control; specificity and compound terms; structure and relationships; auxiliary retrieval devices; multilingual thesauri; AAT Compound Term Rules; and many more. The US ANSI/NISO Z39.19 Thesaurus construction standard is also covered.

Excerpt

The first and second editions of this work were published in 1972 and 1987; a fifteen-year period that saw a burgeoning of computerized databases around the world. The first edition of this work was translated into Portuguese, and the second into Japanese and French; while a Spanish work incorporated a translation into an extended text. Many of the new databases used thesauri, and clearing houses were established to hold copies, or details, of these compilations. The United Nations family had produced over 20 specialist thesauri during this period; most of the large database providers supplied thesauri; and most software suppliers integrated thesauri into their retrieval capabilities.

However, by the time the third edition appeared in 1997, there had been a dramatic change in the text retrieval scene. Full text retrieval had become the norm; the ugly word 'disintermediation' was coined as information providers targeted the far larger market of end-users; new software suppliers came to the market with new techniques, which they claimed obviated the need for such old-fashioned things as thesauri. Above all, the Internet and its use grew at a staggering pace and search engines proliferated. When the authors were approached by the publisher with a suggestion that they prepare a third edition, they experienced some trepidation. The choice lay between a relatively straightforward job of updating and a complete rewrite; considering, for example, the ways in which text retrieval software handled the language problem. Encouraged by the observation of Milstead (Milstead 1995) that thesauri were not obsolete, merely invisible, the authors opted for the update approach.

But, while still largely in agreement with Milstead, there has been a completely new development in the world of retrieval, brought about by the poorly organized glut of information available on the Internet and the growing number of intranets. This development is known as the 'corporate taxonomy', which in its fuller manifestation incorporates elements of both classification and thesauri: both of which may be made visible, in whole or in part, being displayed on the screen or in printed directories. One software supplier introduces the topic on its website with the following words: 'In one of the ironic twists of modern technology, the widespread acceptance of the Internet has led to a resurgence of interest in the much older, traditional study of knowledge classification. While the classification and categorization of information previously rested primarily in academia and the library sciences, today any large website or corporation finds itself facing the problem of how to organize and structure the information it provides to its users. The problem isn't a trivial one; Yahoo! employs hundreds of “surfers” to maintain its web directory, while corporations employ knowledge management architects to maintain corporate intellectual capital sites. The Internet may have automated the distribution of information, but it has created a new army of information categorizers doing manual, time-consuming work'. This same supplier (of automatic categorization software) goes on to suggest an architecture comprising the three layers of Ontology, Taxonomy and Thesaurus. An Ontology is . . .

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.