This paper presents our system and results for the Feed Distillation task in the Blog track at TREC 2007. Our experiments focus on two dimensions of the task: (1) a large-document model (feed retrieval) vs. a small-document model (entry or post retrieval) and (2) a novel query expansion method using the link structure and link text found within Wikipedia.
Jonathan L. Elsas, Jaime Arguello, Jamie Callan, J