Automatic and multilingual documentary representation of technical-scientific texts (SISTRA)

• Grant number: 19/25470-6
• Support Opportunities: Regular Research Grants
• Duration: November 01, 2020 - October 31, 2022
• Field of knowledge: Applied Social Sciences - Information Science - Library and Information Science
• Convênio/Acordo: CONFAP - National Council of State Research Support Foundations
• Principal Investigator: Mariângela Spotti Lopes Fujita
• Grantee: Mariângela Spotti Lopes Fujita
• Host Institution: Faculdade de Filosofia e Ciências (FFC). Universidade Estadual Paulista (UNESP). Campus de Marília. Marília , SP, Brazil
• Associated researchers: Isidoro Gil Leiva
• Associated scholarship(s): 21/01699-4 - Development and implementation of the SIStema for the Automatic Processing of a multilingual e-book for technical-scientific texts - SISTRA, BP.TT

Abstract

Traditionally, university libraries have increased their collections through the purchase or individual exchange of printed books, which has made it possible to properly apply the usual cataloging techniques for embedded copies. With the emergence of the electronic book, the processes of book entry into libraries have gradually changed. Nowadays, university libraries, in addition to purchasing individual ebooks, allowing us to continue cataloging each copy conveniently to make them available to users, have incorporated new and varied forms of ebook entry, such as buying or subscribing to e-books. ebooks that can be thousands of ebooks at once. This massive embedding of ebooks makes it impossible to properly catalog each copy because of the time needed to make them accessible to users on the query platform. Therefore, these changes in the way ebooks have been acquired has meant that what we understand as traditional cataloging is performed only on certain ebooks based on certain criteria such as certain subjects or that ebooks are recommended by teachers as a mandatory reference resource. Therefore, almost all catalog e-books or search platforms currently available to university library users contain descriptive data provided by e-book vendors. Cataloging data provided by eBook providers generally does not provide cataloging by subject or is too general a subject or sometimes only by publishers but not by university libraries. This causes the search options to be significantly reduced, which will have a negative impact on users. In this context, of changes in the technical processes that are taking place in university libraries, this research project is born. Libraries need to catalog subject matter so that document representation meets the needs of retrieving information on the web through their online catalogs. These information retrieval needs in online catalogs are increasingly specific and specialized according to the profile of users who access them through the internet from anywhere. On the other hand, in order to make it easier for library catalogers to use subject cataloging procedures and controlled vocabularies to represent more specifically and exhaustively documented book contents, it is important to use tools available in an automated system that ensures quality uniformity. Therefore, this project proposes the development of methods and tools for automatic and semi-automatic representation of ebooks for libraries to obtain good recovery by subject of their online catalogs and metabuscators; provide theoretical and methodological support for the elaboration, implementation and evaluation of methods and tools of automatic and semi-automatic representation in library environments; Provide the formation of the indexer with sufficient knowledge and practice of a consistent methodology for the elaboration, evaluation and control of methods and tools of automatic and semi-automatic representation, enabling him/her in Information Science and interdisciplinary areas methodologies; Contribute to the curricular development of syllabus of the Information Organization disciplines of Undergraduate Librarianship Courses and indexing training courses for in-service training of information professionals; disseminate the importance and influence that automatic and semi-automatic indexing methods and tools have on the performance of retrieving information by library subject matter; and raise awareness and empower those responsible for indexing policy: library leaders and catalogers. (AU)