Extracting and exploring the geo-temporal semantics of textual resources

Bruno Martins 0001; Hugo Manguinhas; José Borbinha; Hugo Miguel Álvaro

Informações chave

Autores:

Bruno Martins 0001 (Bruno Emanuel Da Graça Martins); Hugo Manguinhas; José Borbinha (José Luís Brinquete Borbinha); Hugo Miguel Álvaro (Hugo Miguel Álvaro Manguinhas)

Publicado em

25/09/2008

Resumo

Geographical and temporal criteria both have important roles in filtering, grouping and prioritizing information resources. This paper presents techniques for extracting semantic geo-temporal information from textual content, using relatively simple natural language processing methods that leverage on a Web gazetteer service. A prototype system, implementing the proposed methods and capable of displaying the extracted information over maps and timelines, is described in detail. This prototype supports the RSS format as input to the extraction process, demonstrating the possible application of these techniques to content from nearly all online news sites and blogs. Experimental results demonstrate the efficiency and accuracy of the proposed approaches.