Problems of Subject Retrieval in the Electronic Catalogue of Scientific Library: Review of Studies

Abstract

Scientific libraries use different approaches and tools for subject retrieval. Issues related to electronic subject retrieval are widely discussed in the library community. The purpose of this review is to analyse publications of recent years on the aspects and problems related to subject retrieval in electronic catalogue (EC). For its preparation, the author used publications (and references to them) taken from the national bibliographic database — the Russian Science Citation Index (RSCI); the time coverage is 2003—2019. Particular attention in the studies is focused on the need to combine different information retrieval languages (IRL), thesauruses, subject headings and keywords to ensure effective and multi-aspect search. Since the domestic libraries have accumulated a lot of experience in indexing and retrieval through hierarchical classifications (UDC and LBC), it is necessary to use this at the full extent in the electronic subject retrieval. The authors of the recent studies on subject retrieval consider being the most effective the functioning of classification systems as part of the complex of linguistic tools of electronic catalogue. In the electronic environment, it becomes possible to use classification indexes as a set of characters that maximally reflect the content of the document, what involves the use of all fragments of the index as independent retrieval elements. However, the lexical composition of the classification information retrieval languages needs to be modified for effective use in the electronic catalogue. The main directions here can be as follows: unification of the structure and vocabulary of classification schedules, optimization of the structure of classification systems (using facetization principles), as well as formalization of the use of grammar tools, (it is necessary to ensure the unambiguity of their interpretation in the electronic catalogue). The review presents and analyses the areas where the subject retrieval has been developed in the electronic environment for over the past 16 years. For example, there has been developed and is being implemented the superstructure over Universal Decimal Classification (Library for Natural Sciences of the Russian Academy of Sciences), which allows users of the library to remotely conduct full-fledged subject retrieval. The paper also presents the new developments by the Russian State Library on subject retrieval, in particular, the project “Presentation of classification metadata of electronic libraries using linked data technology”. The author concludes that it is necessary to form a common space of related open data of various libraries.