Systems for automatic indexing by assignment: a comparative analysis

Objective: This work presents a comparative analysis between two multilingual automatic indexing systems that perform term assignment: SISA and MAUI. The SISA (Semi-automatic Indexing System) made in Spain and initially proposed for the area of Librarianship and Documentation, it is a specialist sys...

Descripción completa

Detalles Bibliográficos
Autores: Silva, Sâmela Rouse de Brito, Correa, Renato Fernandes
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2020
País:Brasil
Institución:Universidade Federal de Santa Catarina (UFSC)
Repositorio:Encontros Bibli
Idioma:portugués
OAI Identifier:oai:periodicos.ufsc.br:article/70740
Acceso en línea:https://periodicos.ufsc.br/index.php/eb/article/view/1518-2924.2020.e70740
Access Level:acceso abierto
Palabra clave:Indexação Automática
Indexação Automática por Atribuição
Sistema de Indexação Automática
Processamento de Linguagem Natural
Recuperação da Informação
Automatic Indexing
Automatic Indexing by Assignment
Automatic Indexing Systems
Natural Language Processing
Information retrieval
Descripción
Sumario:Objective: This work presents a comparative analysis between two multilingual automatic indexing systems that perform term assignment: SISA and MAUI. The SISA (Semi-automatic Indexing System) made in Spain and initially proposed for the area of Librarianship and Documentation, it is a specialist system that automatically indexes following a thesaurus and predetermined rules of indexation which are based on the frequency and position of the terms. The MAUI (Multi-purpose Automatic Topic Indexing) is a system of New Zealand origin that presents the specificity of use of a thesaurus and algorithm of machine learning to generate model through the results of the intellectual indexing, being the terms represented by statistical features. Methods: The research is exploratory and bibliographical, where the method used to construct this work was the comparative study based on content analysis of the scientific publications containing experience reports of application of that software. The stages of the research consisted of describing and comparing the characteristics of each system, raising information about how the documents are processed, how the systems performs the extraction and selection of the descriptors terms, and the application context. Results: The results show the   approaches, main operations, the resources used by each system during the automatic indexing process, as well as the application context and quality of results. Conclusions: It hopes to contribute to the studies on the topic of automatic indexing in the deepening discussion about descriptive and comparative categories related to methods and techniques implemented in the systems.