Check the relevance of the description made. Establish a hierarchical enumeration of the indexing terms according to their importance within the analyzed text. Formalize the description according to the presentation rules of the system.
Store, on the computer, the information related to the headings based on the instructions of the LILDBI-WEB application. At the end of the synthetic analytical processing, in the case of books lacking an abstract, postpone filling in this application until the preparation of the abstract is completed. 14.4 Documented Procedure The indexing process comprises several stages, differentiated according to the various opinions of different authors. According to our system, it is subject to 3: examination of the document and determination of its content, identification and selection of the main content concepts and selection of indexing terms with a view to analyzing the semantic content of a document to delimit the concepts essential or pertinent that make it possible to characterize it, synthetically, by means of the most appropriate indices (terms). Although requests for information are also subject to analysis by indexing, they are ignored in this context because they are a description of technical processes and not services. Serial publications from admission and registration are submitted to this process (indexing), as well as the originals of the articles included in Cuban journals published by ECIMED. As for books, brochures and theses, the synthetic analytical processing started with cataloging and classification continues. In the BMN, books, brochures and theses are indexed at a monographic level, while the analytical level applies both to articles of foreign serials –not processed by MEDLINE, previously selected and approved according to its collection development policy–, as well as articles from Cuban magazines and book chapters of Cuban origin that comply with the requirements established in the guide for the selection of documents (LILACS methodology).
To carry out this process, the collection development policy must be considered. In relation to recently published Cuban journals, they are indexed at the analytical level to add to the LILACS database. The printed annotated DeCS is used, mainly, as well as the automated DeCS contained in LILDBI, when conditions allow it. The mechacopists enter all the information related to each record in the LILDBI application, reproduce, from the published journal, the data corresponding to the bibliographic description and the main descriptors and then obtain the secondary and precoded descriptors from an auxiliary file where the computer -Librarians reflect that data.