Šrndić N. Machine learning and security of non-executable files

University of Tübingen, 2017. — 148 p.Computer malware is a well-known threat in security which, despite the enormous time and effort invested in fighting it, is today more prevalent than ever. Recent years have brought a surge in one particular type: malware embedded in non-executable file formats, e.g., PDF, SWF and various office file formats. The result has been a massive number of infections, owed primarily to the trust that ordinary computer users have in these file formats. In addition, their feature-richness and implementation complexity have created enormous attack surfaces in widely deployed client software, resulting in regular discoveries of new vulnerabilities.