This paper presents the work done for the TREC 2008 blog distillation task. We introduce two new methods based on blog site search using resource selection which was the framework we used for the TREC 2007 blog distillation task. One is a new factor that penalizes the topical diversity of a blog. The other is a query expansion technique. We compare the methods to strong baselines.
Jangwon Seo, W. Bruce Croft