PROPOSED PROCEDURE FOR THE SEMI-AUTOMATIC CONSTRUCTION OF THESAURI IN UNIVERSITY LIBRARIES

Objective Theoretical procedures are proposed in indexing, and at the same time being introduced in plots of Computer Science to solve problems of the ICT Network of the VLIR Project from a multidisciplinary perspective, given the characteristics they impose the challenge of trying to develop any normative process in the field of indexing for a complex and multidisciplinary environment, being the first time that a job in Information Sciences is assumed at the Central University "Marta Abreu" of Las Villas (UCLV). Design / Methodology / Approach the techniques of mathematical origin were used: Zip's Law, TF-IDF, N-grams and Stop World Elimination, providing a predominantly quantitative mixed approach. The description serves as a guide for the construction of specialized lexicons, by including the construction mechanisms based on the internationally required rules. Results / Discussion: From the diagnosis of the Indexing Processes in the Libraries associated to the TIC Network of the VLIR Project, methods and techniques were applied in the collection of information that allowed obtaining results related to the experience in the indexing and thesaurus construction processes. in the various universities of the country. Conclusions: The creation of a procedure that provides the transformation of controlled vocabularies into an interoperable language, facilitates indexing and efficient information retrieval. Originality / Value: The social impact of use is that by containing standardized data in SKOS format, the platforms used and developed by the network will be able to interpret with other platforms with similar purposes, giving visibility to the science of the ICT Network.