Proactive Detection of Unknown Binary Executable Malware
Éric Filiol · InTech eBooks · 2017
Not cached yet. This record was fetched live for this request and is stored after the lookup. Reopening the page should serve the cached copy.
InTech eBooks · OpenAlex · book-chapter · 2017
Abstract
To detect unknown malware, heuristic methods or more generally statistical approaches are the most promising research trends nowadays, but their computing and detection performances are generally not compatible with what users do accept. Hence, most commercial AV products still heavily rely on signature-based detection (opcodes, control flow graph, and so on). This implies that frequent and prior updates must be performed. May their analysis techniques be fully static of dynamic (using sandboxing or virtual machines), commercial AVs do not capture what defines malware compared to benign files: their intrinsic actions.
In this chapter, we focus on binary executables and we describe how to effectively synthetize these actions and what are the differences between malware and nonmalicious files. We extract and analyze two tables that are present in executable files: the import address table (IAT) and export address table (EAT). These tables summarize the different interactions of the executable with the operating system. We show how this information can be used in supervised learning to provide effective detection algorithms, which have proven to be very accurate and proactive with respect to unknown malware detection.
Open the resolved locator · 10.5772/67775
Metadata licence: OpenAlex metadata (CC0)