Automatic Extraction of Quotes and Topics from News Feeds

• Author(s): Luís António Diniz Fernandes de Morais Sarmento ; Sérgio Sobral Nunes
• Date: 2009
• Persistent ID: http://hdl.handle.net/10216/7080
• Origin: Repositório Aberto da Universidade do Porto
• Subject(s): Ciências sociais; Ciências da comunicação; Serviços de informação na Internet, Ciências tecnológicas; Tecnologia; Tecnologia da comunicação, Humanidades; Ciência da informação; Gestão de informação; Processamento de informação, Humanidades; Ciências da linguagem; Linguística; Linguística computacional
• Document Type: Conference Object
• Language: English

Description

The explosive growth in information production poses increasing challenges to consumers, confronted with problems often described as "information overflow". We present verbatim, a software system that can be used as a personal information butler to help structure and filter information. We address a small part of the information landscape, namely quotes extraction from portuguese news. This problem includes several challenges, specifically in the areas of information extraction and topic distillation. We present a full description of the problems and our adopted approach. verbatim is available online at http://irlab.fe.up.pt/p/verbatim.