FEUP at TREC 2010 blog track: Using h-index for blog ranking

• Autor(a) principal: José Devezas
• Data de Publicação: 2010
• Outros Autores: Sérgio Nunes, Cristina Ribeiro
• Tipo de documento: Livro
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: https://hdl.handle.net/10216/66281
• topic: Tecnologia da informação, Engenharia de computadores, Ciências da computação e da informação; Information technology, Computer engineering, Computer and information sciences

This paper describes the participation of FEUP, from the University of Porto, in the TREC 2010 Blog Track. FEUP participated in the baseline blog distillation task with work focused on the use of link features available in the TREC Blogs08 collection. The approach presented in this paper uses the link information available in most individual posts to amplify each post's score. Blog scores, and subsequent ranks, are obtained by combining individual post scores. We boost post scores using the in-degree of each post and the h-index of each blog. This results in an improvement of P@10, over our baseline, for the in-degree and the h-index runs. When compared to the in- degree, the h-index run results in higher performance values for each of the applied evaluation metrics.