Content area
As our world becomes increasingly dependent on technology, the advancement of digital forensics has become a key focus in the fight against cybercrime. The forensic community depends on the availability of disk images, network captures, and other forensic artifacts for education, tool validation, and research. However, real-world datasets often contain sensitive information that may be difficult to remove, making them challenging to distribute publicly. As a result, researchers and educators can encounter gaps in available datasets, typically leading to the manual development of new datasets. While viable, this approach is time-consuming and rarely produces datasets that accurately reflect real-world scenarios suitable for comprehensive training and education. In turn, there is ongoing research into forensic synthesizers, which automate the process of creating unique, synthetic datasets that can be publicly distributed without legal and other logistical concerns.
This thesis introduces the automated kinetic framework, or AKF, a modular synthesizer for creating and interacting with virtualized environments to simulate human activity. AKF significantly improves upon the designs and implementations of prior synthesizers while largely maintaining feature parity and usability. Additionally, AKF leverages the CASE standard to provide human- and machine-readable reporting, exposing low-level dataset features in a searchable format. Finally, this thesis describes options for leveraging generative AI to develop high-level scenarios as well as individual artifacts. These contributions are intended to improve the speed at which synthetic datasets can be created and ensure the long-term usefulness of AKF-generated datasets and the framework as a whole.