The increasing popularity of Twitter renders improved trustworthiness and relevance assessment of tweets critical for search. However, given the limitations on the size of tweets, it is hard to extract measures for ranking from the tweets' content alone. We propose a method of ranking tweets by generating a Feature Score for each tweet that is based not just on content, but also additional information from the Twitter ecosystem that consists of users, tweets, and the webpages that the tweets link to. The Feature Score is propagated over an agreement graph based on tweets' content similarity. The propagated Feature Score that is sensitive to content popularity and trustworthiness is used to rank the tweets for a query. An evaluation of our method on 16 million tweets from the TREC 2011 Microblog Dataset shows that it doubles the precision over the baseline Twitter Search, and outperforms the best-performing method on the TREC 2011 Microblog dataset.
展开▼
