Query Expansion for Blog Searching
A brief description of some recent (and partly still ongoing) experiments on query expansion/query modeling.
Project
When retrieval systems fail, they often fail because they do not uncover important aspects of a query. To help uncover aspects of a query, we examine the use of query expansion against multiple external corpora. We consider two informal text retrieval tasks: blog post finding and blog finding. By combining perspectives (i.e., query expansion terms) found in the blog corpus with perspectives identified in a knowledge source or in a news corpus, we identify additional aspects of the search topic.
The experiments make use of three corpora:
- The TRECBlog06 corpus
- The AQUAINT-2 corpus
- A Wikipedia dump of April 2008
On top of these we use Indri/Lemur to expand our queries and do the initial and final retrieval.
Project Partners
- University of Amsterdam
