Meu SciELO
Serviços Personalizados
Artigo
Indicadores
Citado por SciELO
Links relacionados
Similares em SciELO
Compartilhar
Revista Cubana de Ciencias Informáticas
versão On-line ISSN 2227-1899
Resumo
TORRES LOPEZ, Carmen e ARCO GARCIA, Leticia. Textual representation in semantic vector space. Rev cuba cienc informat [online]. 2016, vol.10, n.2, pp. 148-180. ISSN 2227-1899.
The vector space model represents textual documents via vectors of terms, but it cannot represent semantic relationships between words. Semantic vector spaces are based on the idea that the meaning of a word can be learned from a linguistic environment and have two approaches, the distributional semantics and compositional semantics. The first approach analyzes the meaning of individual words and the second approach the meaning of phrases, sentences and paragraphs. This review presents the main models of these approaches and the computational tools that implement them. This study bring to a conclusion that the incorporation of semantic representations in the different tools that perform textual analysis is necessary, essentially researchers have obtained best representations that make prediction of contexts in the case of distributional models and the ones that incorporate models based on neural networks for compositional models.
Palavras-chave : text mining; semantic vector space; distributional semantics; compositional semantics.