Visual Analytics : Foundations and Experiences in Malware Analysis

Abstract

This chapter starts by providing some background in behavior-based malware analysis. Subsequently, it introduces VA and its main components based on the knowledge generation model for VA (Sacha et al., 2014). Then, it demonstrates the applicability of VA in in this subfield of software security with three projects that illustrate practical experience of VA methods: MalwareVis (Zhuo et al., 2012) supports network forensics and malware analysis by visually assessing TCP and DNS network streams. SEEM (Gove et al., 2014) allows visual comparison of multiple large attribute sets of malware samples, thereby enabling bulk classification. KAMAS (Wagner et al. 2017) is a knowledge-assisted visualization system for behavior-based malware forensics enabled by API calls and system call traces. Future directions in visual analytics for malware analysis conclude the chapter.