skip to main content

Thesauri are treasuries of words that share alphabetical order with dictionaries, but also present a conceptual network structure between terms of one or more fields of human knowledge. This conceptual order, consisting of various kinds of links between words, has many usages in information organization and retrieval, information analysis and scientometrics, educational planning, and the like. It is also an invaluable key prerequisite for building ontologies as tools for the Semantic Web. Thesauri organize words, and concepts -expressed by terms- are arranged in a manner that clearly illustrates the links between them. For instance, preferred terms are accompanied by entries for their synonyms and near-synonyms. Thesauri guide both the indexer and the user to choose similar term or similar combinations of preferred terms to represent a given subject. Standardizing terms for a subject leads to uniform entries in indexing. If homonyms are correctly separated so that each one points only to a single meaning, and the preferred terms for indexing are selected from a set of synonyms and near-synonyms, the indexer and the user will both find similar or near-similar terms for the same concept.

IRANDOC began working on thesauri in 1971 with the Handbook of Thesauri Building Conventions and Principles ,and in 1996 published its first thesaurus, the System for Transaction of Sci-Tech Information (Nama). From 1997 to 2007, the Inclusive Bilingual Persian-English Thesauri were designed and written on multiple subjects including science and technology fields like mathematics, physics, chemistry, biosciences, earth sciences, engineering and agriculture, covering over 100,000 terms.

By summer 2019, more than 100,000 words were added to the previous set of words, with the semantic relations between them expanded and improved, thus the release of the second version of the thesauri system for public access.

The result of IRANDOC services are presented in the form of the sixteen thesauri covering around 210,000 terms (descriptor and non-descriptor).