Email Archive Analysis Through Graphical Visualization

Li, Wei-Jen; Hershkop, Shlomo; Stolfo, Salvatore

The analysis of the vast storehouse of email content accumulated or produced by individual users has received relatively little attention other than for specific tasks such as spam and virus filtering. Current email analysis in standard client applications consists of keyword based matching techniques for filtering and expert driven manual exploration of email files. We have implemented a tool, called the Email Mining Toolkit (EMT) for analyzing email archives which includes a graphical display to explore relationships between users and groups of email users. The chronological flow of an email message can be analyzed by EMT. Our design goal is to embed the technology into standard email clients, such as Outlook, revealing far more information about a user's own email history than is otherwise now possible. In this paper we detail the visualization techniques implemented in EMT. We show the utility of these tools and underlying models for detecting email misuse such as viral propagation, and spam spread as examples.


More About This Work

Academic Units
Computer Science
Published Here
April 30, 2010


VizSEC/DMSEC '04: proceedings of the 2004 ACM Workshop on Visualization and Data Mining for Computer Security, Washington, DC, USA, October 29, 2004, co-located with CCS 2004 (New York: ACM, 2004), pp. 128-132.