posted on 2012-11-01, 00:00authored byYubin Kim, Reyyan Yeniterzi, Jamie Callan
One major difficulty in performing ad-hoc search on microblogs such as Twitter is the limited vocabulary of each document due their short length. In this paper, two approaches to addressing this issue are presented. The first is query expansion through pseudo-relevance feedback and the other is document expansion of tweets using web documents linked from the body of the tweet. Tweets are expanded by concatenating the contents of the title tag and the meta descriptor tags of the document to the tweet itself. These two approaches gave additive gains in MAP and Precision at 30.