| Kurzfassung
|
As data becomes increasingly digital, the impact of compromised data can be disastrous. To mitigate these risks, various threat modeling approaches and taxonomies have been developed. However, many of these approaches still require considerable effort and expertise. This thesis presents an approach that extends the architecture-based data flow analysis framework xDECAF, enabling software architects to easily include semantic information in Data Flow Diagrams (DFDs). To automatically identify threats within these DFDs, we defined an initial set of constraints based on the mature privacy threat taxonomy LINDDUN. A user study suggests that this approach can significantly decrease the time required for analysis compared to LINDDUN Pro, but the translation seems to be subpar.
|
