Information Leaks and Suggestions: A Case Study using Mozilla Thunderbird
Vitor Rocha de Carvalho, Ramnath Balasubramanyan, William W. Cohen · 2009
People often make serious mistakes when addressing email messages. One type of costly mistake is an“email leak”, i.e., accidentally sending a message to an unintended recipient — a widespread problem that can severely harm individuals and corporations. Another type of addressing error is forgetting to add an intended collaborator as recipient, a likely source of costly misunderstandings and communication delays in large corporations. To address these problems, various data mining techniques have been proposed recently [3, 4]. In this paper we describe the deployment of some of these techniques in a popular email client ( Mozilla Thunderbird ), and report how users responded to such data mining techniques in their everyday lives. In spite of interface, privacy and speed constraints, results were fairly positive. More than 15% of the users reported that the client prevented real cases of email leaks, and more than 47% of them accepted recommendations provided by the data mining techniques. We then conclude by presenting a few lessons learned from this deployment, and discussing costs and benefits of making these techniques permanent additions to email clients.