Busca avançada
Ano de início
Entree


Text characterization based on recurrence networks

Texto completo
Autor(es):
Souza, Barbara C. E. ; Silva, Filipi N. ; de Arruda, Henrique F. ; da Silva, Giovana D. ; Costa, Luciano Da F. ; Amancio, Diego R.
Número total de Autores: 6
Tipo de documento: Artigo Científico
Fonte: INFORMATION SCIENCES; v. 641, p. 15-pg., 2023-05-12.
Resumo

Several complex systems are characterized by exhibiting intricate properties that occur at multiple scales. These multi-scale characterizations are used in various applications. In particular, texts can be characterized by a hierarchical structure, which can be approached by using multi-scale concepts and methods. Here, we adopt an extension of the multi-scale, mesoscopic approach - hereafter referred to as a recurrence network - to represent text narratives, in which only the recurrent relationships among tagged parts of speech (subject, verb and direct object) are considered to establish connections among sequential pieces of text. The characterization of the texts was then achieved by considering scale-dependent complementary methods: accessibility and symmetry. To evaluate the potential of these concepts, we approached the problem of distinguishing between meaningful and meaningless texts and different literary genres (namely, fiction and non-fiction). A set of 300 books was considered and compared by using the above approaches. The recurrence network characterization was able to discriminate to some extent between real and meaningless and between the two genres assessed. Thus, our results indicate that recurrence networks are able to capture subtleties in book plots, suggesting that a similar methodology can be used in related networked applications. (AU)

Processo FAPESP: 19/07665-4 - Centro de Inteligência Artificial
Beneficiário:Fabio Gagliardi Cozman
Modalidade de apoio: Auxílio à Pesquisa - Programa eScience e Data Science - Centros de Pesquisa em Engenharia
Processo FAPESP: 21/01744-0 - Usando redes complexas para caracterizar e identificar o sucesso de obras
Beneficiário:Giovana Daniele da Silva
Modalidade de apoio: Bolsas no Brasil - Iniciação Científica
Processo FAPESP: 20/06271-0 - Combinando redes complexas e word embeddings em tarefas de classificação de textos
Beneficiário:Diego Raphael Amancio
Modalidade de apoio: Auxílio à Pesquisa - Regular
Processo FAPESP: 18/10489-0 - Transformações de redes complexas e suas implicações na topologia e dinâmica de sistemas complexos
Beneficiário:Henrique Ferraz de Arruda
Modalidade de apoio: Bolsas no Brasil - Pós-Doutorado
Processo FAPESP: 15/22308-2 - Representações intermediárias em Ciência Computacional para descoberta de conhecimento
Beneficiário:Roberto Marcondes Cesar Junior
Modalidade de apoio: Auxílio à Pesquisa - Temático