• Medientyp: Elektronische Hochschulschrift; E-Book; Dissertation; Sonstige Veröffentlichung
  • Titel: Defending Against Targeted Attacks with Pattern Recognition ; Mustererkennungsbasierte Verteidgung gegen gezielte Angriffe
  • Beteiligte: Gascón Polanco, Hugo [VerfasserIn]
  • Erschienen: TU Braunschweig: LeoPARD - Publications And Research Data, 2020-02-06
  • Umfang: 140 Seiten
  • Sprache: Englisch
  • DOI: https://doi.org/10.24355/dbbs.084-202002060816-0
  • Schlagwörter: doctoral thesis
  • Entstehung:
  • Anmerkungen: Diese Datenquelle enthält auch Bestandsnachweise, die nicht zu einem Volltext führen.
  • Beschreibung: The speed at which everything and everyone is being connected considerably outstrips the rate at which effective security mechanisms are introduced to protect them. This has created an opportunity for resourceful threat actors which have specialized in conducting low-volume persistent attacks through sophisticated techniques that are tailored to specific valuable targets. Consequently, traditional approaches are rendered ineffective against targeted attacks, creating an acute need for innovative defense mechanisms. This thesis aims at supporting the security practitioner in bridging this gap by introducing a holistic strategy against targeted attacks that addresses key challenges encountered during the phases of detection, analysis and response. The structure of this thesis is therefore aligned to these three phases, with each one of its central chapters taking on a particular problem and proposing a solution built on a strong foundation on pattern recognition and machine learning. In particular, we propose a detection approach that, in the absence of additional authentication mechanisms, allows to identify spear-phishing emails without relying on their content. Next, we introduce an analysis approach for malware triage based on the structural characterization of malicious code. Finally, we introduce MANTIS, an open-source platform for authoring, sharing and collecting threat intelligence, whose data model is based on an innovative unified representation for threat intelligence standards based on attributed graphs. As a whole, these ideas open new avenues for research on defense mechanisms and represent an attempt to counteract the imbalance between resourceful actors and society at large. ; In unserer heutigen Welt sind alle und alles miteinander vernetzt. Dies bietet mächtigen Angreifern die Möglichkeit, komplexe Verfahren zu entwickeln, die auf spezifische Ziele angepasst sind. Traditionelle Ansätze zur Bekämpfung solcher Angriffe werden damit ineffektiv, was die Entwicklung innovativer Methoden unabdingbar ...
  • Zugangsstatus: Freier Zugang