News topic tracking and re-ranking with query expansion based on near-duplicate detection

Abstract

Increase of digital storage capacity enabled the creation of large-scale news video archives. To make full use of the archive, it is necessary to grasp the development and dependencies of news stories. Considering this problem, we investigate tracking and re-ranking methodologies of news stories. The archive used as a test-bed consists of more than 30,000 news stories. This paper proposes a novel scheme of mining topic-related stories through a query-expansion algorithm on the basis of near duplicates built on top of text. Experiments showed that the query-expansion algorithm based on near-duplicate constraints outperformed traditional methods that only use textual features. © 2009 Springer-Verlag Berlin Heidelberg.