Finding Relevant Tweets

When a user of a microblogging site authors a microblog<br/>post or browses through a microblog post, it provides cues as to what<br/>topic she is interested in at that point in time. Example-based search<br/>that retrieves similar tweets given one exemplary tweet, such as the one<br/>just authored, can help provide the user with relevant content. We investigate<br/>various components of microblog posts, such as the associated<br/>timestamp, author’s social network, and the content of the post, and<br/>develop approaches that harness such factors in finding relevant tweets<br/>given a query tweet. An empirical analysis of such techniques on real<br/>world twitter-data is then presented to quantify the utility of the various<br/>factors in assessing tweet relevance. We observe that content-wise similar<br/>tweets that also contain extra information not already present in the<br/>query, are perceived as useful. We then develop a composite technique<br/>that combines the various approaches by scoring tweets using a dynamic<br/>query-specific linear combination of separate techniques. An empirical<br/>evaluation establishes the effectiveness of the composite technique, and<br/>that it outperforms each of its constituents.<br/>