Abstract
In the micro-blogging service Twitter, the sparseness of text messages is an enormous obstacle in extracting keyphrases from tweets. However, regardless of the sparseness in text, tweets include an abundant number of links in the form of hashtags. This paper investigates the possibility of leveraging hashtags in tweets to enhance the graph-based keyphrase extraction. By using an auxiliary set of tweets found in hashtags, we show that we can improve extracting keyphrases from tweets by augmenting the graph with a wider knowledge context. Specifically, we propose two different approaches for choosing the best hashtags links to use for enhancing graph-based keyphrase extraction by either using a frequency approach or a hybrid approach that uses multiple methods for cleverly choosing the best hashtags. Experiments on the proposed approaches showed an improvement in the range of 9% to 37% over the case of ignoring the hashtag links.