Agrupamento hierárquico incremental de textos com incorporação de entidades nomeadas

Abstract

The increasing use of the internet has generated a large amount of information made available to the users ever more frequently. This scenario can be easily verified in websites of news and social media. With this large amount of data, the need to organize data in an automatic manner emerges, with the intent of providing easy navigation and localization of this information to the users. The hierarchical grouping techniques of text documents aim at organizing texts in many levels of granularity, resulting in a structure that meets such needs. The incremental grouping techniques aim at working with the dynamicity of the data, adapting the grouping structures according to data added or removed. Many of the documents available for analysis can contain names of entities such as people, companies and locations. Such information can be incorporated to the grouping process with the objective of aiding these techniques in their efficiency and document organization. This work had the objective of studying incremental hierarchical grouping techniques, as well as proposing and investigating the use of entities named in the grouping process.