Discovery and regeneration of hidden emails

Abstract

The popularity of email has triggered researchers to look for ways to help users better organize the enormous amount of information stored in their email folders. One challenge that has not been studied extensively in text mining is the reconstruction of hidden emails. A hidden email is an original email that has been quoted in subsequent emails but is not itself present in the folder; it may have been deleted or may never have been received. This paper proposes a method for reconstructing hidden emails using the embedded quotations found in messages further down the thread hierarchy. To do so, we model all the quoted fragments in a precedence graph, from which hidden emails are regenerated as bulletized documents. The bulletized model is our solution to the situation when a total ordering of fragment is not possible. We give a necessary and sufficient condition for each component of the precedence graph to be captured in a single bulletized email, and we develop heuristics that minimize the number of regenerated emails when the condition is not met. Finally, we present empirical results showing the scalability of our approach.