Countering network-level denial of information attacks using information visualization

We are besieged with information every day, our inboxes overflow with spam and our search queries return a great deal of irrelevant information. In most cases there is no malicious intent, just simply too much information. However, if we consider active malicious entities, the picture darkens. Denial of information (DoI) attacks assail the human through their computer system and manifest themselves as attacks that target the human's perceptual, cognitive and motor capabilities. By exploiting these capabilities, attackers reduce our ability to acquire and act upon desired information. Even if a traditional denial of service attack against a machine is not possible, the human utilizing the machine may still succumb to DoI attack. When successful, DoI attacks actively alter our decision making, often without our knowledge. In this dissertation, we address the problem of countering DoI attacks. We begin by presenting a taxonomy and framework of DoI attacks and countermeasures to add structure to the problem space. We then closely examine the use of information visualization as a countermeasure. Information visualization is a powerful technique that taps into the high bandwidth visual recognition capability of the human and is well suited to resist DoI attack. Unfortunately, most information visualization systems are designed without a clear emphasis on protecting the human from malicious activity. To address this issue we present a general framework for information visualization system security analysis. We then delve deeply into countering DoI in the network security domain using carefully crafted information visualization techniques to build a DoI attack resistant security visualization system. By creating such a system, we raise the bar on adversaries who now must cope with visualization enhanced humans in addition to traditional automated intrusion detection systems and text-based analysis tools. We conclude with a human-centric evaluation to demonstrate our system's effectiveness.